Upload full run folder: v0-20260123-013837

v0-20260123-014416/args.json

@@ -0,0 +1,351 @@
+{
+  "output_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416",
+  "overwrite_output_dir": false,
+  "do_train": false,
+  "do_eval": false,
+  "do_predict": false,
+  "eval_strategy": "steps",
+  "prediction_loss_only": false,
+  "per_device_train_batch_size": 16,
+  "per_device_eval_batch_size": 16,
+  "per_gpu_train_batch_size": null,
+  "per_gpu_eval_batch_size": null,
+  "gradient_accumulation_steps": 4,
+  "eval_accumulation_steps": null,
+  "eval_delay": 0,
+  "torch_empty_cache_steps": null,
+  "learning_rate": 5e-06,
+  "weight_decay": 0.01,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "max_grad_norm": 1.0,
+  "num_train_epochs": 1.0,
+  "max_steps": -1,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": null,
+  "warmup_ratio": 0.02,
+  "warmup_steps": 0,
+  "log_level": "passive",
+  "log_level_replica": "warning",
+  "log_on_each_node": true,
+  "logging_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/runs",
+  "logging_strategy": "steps",
+  "logging_first_step": true,
+  "logging_steps": 100,
+  "logging_nan_inf_filter": true,
+  "save_strategy": "steps",
+  "save_steps": 2000.0,
+  "save_total_limit": 40,
+  "save_safetensors": true,
+  "save_on_each_node": false,
+  "save_only_model": false,
+  "restore_callback_states_from_checkpoint": false,
+  "no_cuda": false,
+  "use_cpu": false,
+  "use_mps_device": false,
+  "seed": 42,
+  "data_seed": 42,
+  "jit_mode_eval": false,
+  "use_ipex": false,
+  "bf16": true,
+  "fp16": false,
+  "fp16_opt_level": "O1",
+  "half_precision_backend": "auto",
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "local_rank": -1,
+  "ddp_backend": null,
+  "tpu_num_cores": null,
+  "tpu_metrics_debug": false,
+  "debug": null,
+  "dataloader_drop_last": false,
+  "eval_steps": 2000.0,
+  "dataloader_num_workers": null,
+  "dataloader_prefetch_factor": null,
+  "past_index": -1,
+  "run_name": "step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF",
+  "disable_tqdm": null,
+  "remove_unused_columns": true,
+  "label_names": null,
+  "load_best_model_at_end": false,
+  "metric_for_best_model": "loss",
+  "greater_is_better": false,
+  "ignore_data_skip": false,
+  "fsdp": "",
+  "fsdp_min_num_params": 0,
+  "fsdp_config": null,
+  "fsdp_transformer_layer_cls_to_wrap": null,
+  "accelerator_config": {
+    "dispatch_batches": false
+  },
+  "deepspeed": null,
+  "label_smoothing_factor": 0.0,
+  "optim": "adamw_torch",
+  "optim_args": null,
+  "adafactor": false,
+  "group_by_length": false,
+  "length_column_name": "length",
+  "report_to": [
+    "wandb"
+  ],
+  "ddp_find_unused_parameters": null,
+  "ddp_bucket_cap_mb": null,
+  "ddp_broadcast_buffers": null,
+  "dataloader_pin_memory": true,
+  "dataloader_persistent_workers": false,
+  "skip_memory_metrics": true,
+  "use_legacy_prediction_loop": false,
+  "push_to_hub": false,
+  "resume_from_checkpoint": null,
+  "hub_model_id": null,
+  "hub_strategy": "every_save",
+  "hub_token": null,
+  "hub_private_repo": null,
+  "hub_always_push": false,
+  "gradient_checkpointing": true,
+  "gradient_checkpointing_kwargs": "{\"use_reentrant\": false}",
+  "include_inputs_for_metrics": false,
+  "include_for_metrics": [],
+  "eval_do_concat_batches": true,
+  "fp16_backend": "auto",
+  "evaluation_strategy": "steps",
+  "push_to_hub_model_id": null,
+  "push_to_hub_organization": null,
+  "push_to_hub_token": null,
+  "mp_parameters": "",
+  "auto_find_batch_size": false,
+  "full_determinism": false,
+  "torchdynamo": null,
+  "ray_scope": "last",
+  "ddp_timeout": 18000000,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null,
+  "dispatch_batches": null,
+  "split_batches": null,
+  "include_tokens_per_second": false,
+  "include_num_input_tokens_seen": false,
+  "neftune_noise_alpha": null,
+  "optim_target_modules": null,
+  "batch_eval_metrics": false,
+  "eval_on_start": false,
+  "use_liger_kernel": false,
+  "eval_use_gather_object": false,
+  "average_tokens_across_devices": false,
+  "sortish_sampler": false,
+  "predict_with_generate": false,
+  "generation_max_length": null,
+  "generation_num_beams": null,
+  "generation_config": null,
+  "tuner_backend": "peft",
+  "vit_gradient_checkpointing": null,
+  "router_aux_loss_coef": 0.0,
+  "enable_dft_loss": false,
+  "enable_channel_loss": false,
+  "check_model": true,
+  "acc_strategy": "token",
+  "train_dataloader_shuffle": true,
+  "max_epochs": null,
+  "aligner_lr": 1e-05,
+  "vit_lr": 5e-06,
+  "use_logits_to_keep": null,
+  "ds3_gather_for_generation": true,
+  "resume_only_model": false,
+  "optimizer": null,
+  "loss_type": null,
+  "metric": null,
+  "eval_use_evalscope": false,
+  "eval_dataset": [],
+  "eval_dataset_args": null,
+  "eval_limit": null,
+  "eval_generation_config": null,
+  "extra_eval_args": null,
+  "use_flash_ckpt": false,
+  "use_ray": false,
+  "ray_exp_name": null,
+  "device_groups": null,
+  "model": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "model_type": "step_audio2_mini",
+  "model_revision": null,
+  "task_type": "causal_lm",
+  "torch_dtype": "bfloat16",
+  "attn_impl": null,
+  "new_special_tokens": [],
+  "num_labels": null,
+  "problem_type": null,
+  "rope_scaling": null,
+  "device_map": null,
+  "max_memory": {},
+  "max_model_len": null,
+  "local_repo_path": null,
+  "init_strategy": null,
+  "template": "step_audio2_mini",
+  "system": null,
+  "max_length": 16384,
+  "truncation_strategy": "delete",
+  "max_pixels": null,
+  "agent_template": null,
+  "norm_bbox": null,
+  "use_chat_template": true,
+  "padding_free": false,
+  "padding_side": "right",
+  "loss_scale": "default",
+  "sequence_parallel_size": 1,
+  "response_prefix": null,
+  "template_backend": "swift",
+  "dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_train90.jsonl"
+  ],
+  "val_dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "cached_dataset": [],
+  "cached_val_dataset": [],
+  "split_dataset_ratio": 0.0,
+  "dataset_num_proc": 1,
+  "load_from_cache_file": false,
+  "dataset_shuffle": true,
+  "val_dataset_shuffle": false,
+  "streaming": false,
+  "interleave_prob": null,
+  "stopping_strategy": "first_exhausted",
+  "shuffle_buffer_size": 1000,
+  "download_mode": "reuse_dataset_if_exists",
+  "columns": {},
+  "strict": false,
+  "model_name": null,
+  "model_author": null,
+  "custom_dataset_info": [],
+  "quant_method": null,
+  "quant_bits": null,
+  "hqq_axis": null,
+  "bnb_4bit_compute_dtype": "bfloat16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "max_new_tokens": 64,
+  "temperature": 0.0,
+  "top_k": null,
+  "top_p": null,
+  "repetition_penalty": null,
+  "num_beams": 1,
+  "stream": false,
+  "stop_words": [],
+  "logprobs": false,
+  "top_logprobs": null,
+  "ckpt_dir": null,
+  "lora_modules": [],
+  "train_type": "lora",
+  "adapters": [],
+  "external_plugins": [],
+  "model_kwargs": {},
+  "load_args": false,
+  "load_data_args": false,
+  "packing": false,
+  "packing_length": null,
+  "packing_num_proc": 1,
+  "lazy_tokenize": true,
+  "custom_register_path": [],
+  "use_hf": false,
+  "ignore_args_error": false,
+  "use_swift_lora": false,
+  "freeze_parameters": [],
+  "freeze_parameters_regex": null,
+  "freeze_parameters_ratio": 0.0,
+  "trainable_parameters": [],
+  "trainable_parameters_regex": null,
+  "freeze_llm": true,
+  "freeze_vit": false,
+  "freeze_aligner": false,
+  "target_modules": [
+    "all-linear"
+  ],
+  "target_regex": null,
+  "target_parameters": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "lora_bias": "none",
+  "lora_dtype": null,
+  "lorap_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "lora_ga_batch_size": 2,
+  "lora_ga_iters": 2,
+  "lora_ga_max_length": 1024,
+  "lora_ga_direction": "ArB2r",
+  "lora_ga_scale": "stable",
+  "lora_ga_stable_gamma": 16,
+  "init_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "use_galore": false,
+  "galore_target_modules": null,
+  "galore_rank": 128,
+  "galore_update_proj_gap": 50,
+  "galore_scale": 1.0,
+  "galore_proj_type": "std",
+  "galore_optim_per_parameter": false,
+  "galore_with_embedding": false,
+  "galore_quantization": false,
+  "galore_proj_quant": false,
+  "galore_proj_bits": 4,
+  "galore_proj_group_size": 256,
+  "galore_cos_threshold": 0.4,
+  "galore_gamma_proj": 2,
+  "galore_queue_size": 5,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "lisa_activated_layers": 0,
+  "lisa_step_interval": 20,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "swanlab_token": null,
+  "swanlab_project": null,
+  "swanlab_workspace": null,
+  "swanlab_exp_name": null,
+  "swanlab_lark_webhook_url": null,
+  "swanlab_lark_secret": null,
+  "swanlab_mode": "cloud",
+  "add_version": true,
+  "create_checkpoint_symlink": false,
+  "zero_hpz_partition_size": null,
+  "deepspeed_autotp_size": null,
+  "early_stop_interval": null,
+  "rank": -1,
+  "global_world_size": 1,
+  "local_world_size": 1,
+  "model_suffix": "Step-Audio-2-mini",
+  "model_info": "ModelInfo(model_type='step_audio2_mini', model_dir='/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini', torch_dtype=torch.bfloat16, max_model_len=16384, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=True, config=None, task_type='causal_lm', num_labels=None)",
+  "model_meta": "ModelMeta(model_type='step_audio2_mini', model_groups=[ModelGroup(models=[Model(ms_model_id='stepfun-ai/Step-Audio-2-mini', hf_model_id='stepfun-ai/Step-Audio-2-mini', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='step_audio2_mini', get_function=<function get_model_tokenizer_step_audio2_mini at 0x7f32c017c4c0>, model_arch=MultiModelKeys(arch_name='step_audio2_mini', embedding=None, module_list=None, lm_head=None, q_proj=None, k_proj=None, v_proj=None, o_proj=None, attention=None, mlp=None, down_proj=None, qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None, language_model=['model', 'lm_head'], aligner=['adapter'], vision_tower=['encoder'], generator=[]), architectures=['StepAudio2ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=True, is_reward=False, is_reranker=False, task_type=None, ignore_patterns=None, requires=['transformers==4.53.3', 'torchaudio', 'librosa'], tags=['audio'])",
+  "model_dir": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "_val_dataset_exists": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "hub": "<class 'swift.hub.hub.MSHub'>",
+  "training_args": "Seq2SeqTrainingArguments(output_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=16, per_device_eval_batch_size=16, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=4, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=5e-06, weight_decay=0.01, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=1.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.02, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=100, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=2000, save_total_limit=40, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=2000, dataloader_num_workers=1, dataloader_prefetch_factor=10, past_index=-1, run_name='step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['wandb'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs={'use_reentrant': False}, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy='steps', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, eval_use_gather_object=False, average_tokens_across_devices=None, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=1e-05, vit_lr=5e-06, use_logits_to_keep=None, ds3_gather_for_generation=True, resume_only_model=False, optimizer='multimodal', loss_type=None, metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='lora', local_repo_path=None, galore_config=None, padding_side='right', padding_free=False, task_type='causal_lm', problem_type=None)"
+}

v0-20260123-014416/checkpoint-2000/README.md

@@ -0,0 +1,202 @@
+---
+base_model: /data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.11.1

v0-20260123-014416/checkpoint-2000/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [],
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "^(encoder.*\\.(query|mlp.0|mlp.2|out|value|key)|adapter.*\\.(linear1|linear2))$",
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

v0-20260123-014416/checkpoint-2000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be7279fd3005dd01721976209db91a2a111273879d3a23d48b36ec7a306aed3e
+size 11988912

v0-20260123-014416/checkpoint-2000/additional_config.json

@@ -0,0 +1 @@
+{"lora_dtype": null, "lorap_lr_ratio": null, "lorap_emb_lr": 1e-06}

v0-20260123-014416/checkpoint-2000/args.json

@@ -0,0 +1,351 @@
+{
+  "output_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416",
+  "overwrite_output_dir": false,
+  "do_train": false,
+  "do_eval": false,
+  "do_predict": false,
+  "eval_strategy": "steps",
+  "prediction_loss_only": false,
+  "per_device_train_batch_size": 16,
+  "per_device_eval_batch_size": 16,
+  "per_gpu_train_batch_size": null,
+  "per_gpu_eval_batch_size": null,
+  "gradient_accumulation_steps": 4,
+  "eval_accumulation_steps": null,
+  "eval_delay": 0,
+  "torch_empty_cache_steps": null,
+  "learning_rate": 5e-06,
+  "weight_decay": 0.01,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "max_grad_norm": 1.0,
+  "num_train_epochs": 1.0,
+  "max_steps": -1,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": null,
+  "warmup_ratio": 0.02,
+  "warmup_steps": 0,
+  "log_level": "passive",
+  "log_level_replica": "warning",
+  "log_on_each_node": true,
+  "logging_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/runs",
+  "logging_strategy": "steps",
+  "logging_first_step": true,
+  "logging_steps": 100,
+  "logging_nan_inf_filter": true,
+  "save_strategy": "steps",
+  "save_steps": 2000.0,
+  "save_total_limit": 40,
+  "save_safetensors": true,
+  "save_on_each_node": false,
+  "save_only_model": false,
+  "restore_callback_states_from_checkpoint": false,
+  "no_cuda": false,
+  "use_cpu": false,
+  "use_mps_device": false,
+  "seed": 42,
+  "data_seed": 42,
+  "jit_mode_eval": false,
+  "use_ipex": false,
+  "bf16": true,
+  "fp16": false,
+  "fp16_opt_level": "O1",
+  "half_precision_backend": "auto",
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "local_rank": -1,
+  "ddp_backend": null,
+  "tpu_num_cores": null,
+  "tpu_metrics_debug": false,
+  "debug": null,
+  "dataloader_drop_last": false,
+  "eval_steps": 2000.0,
+  "dataloader_num_workers": null,
+  "dataloader_prefetch_factor": null,
+  "past_index": -1,
+  "run_name": "step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF",
+  "disable_tqdm": null,
+  "remove_unused_columns": true,
+  "label_names": null,
+  "load_best_model_at_end": false,
+  "metric_for_best_model": "loss",
+  "greater_is_better": false,
+  "ignore_data_skip": false,
+  "fsdp": "",
+  "fsdp_min_num_params": 0,
+  "fsdp_config": null,
+  "fsdp_transformer_layer_cls_to_wrap": null,
+  "accelerator_config": {
+    "dispatch_batches": false
+  },
+  "deepspeed": null,
+  "label_smoothing_factor": 0.0,
+  "optim": "adamw_torch",
+  "optim_args": null,
+  "adafactor": false,
+  "group_by_length": false,
+  "length_column_name": "length",
+  "report_to": [
+    "wandb"
+  ],
+  "ddp_find_unused_parameters": null,
+  "ddp_bucket_cap_mb": null,
+  "ddp_broadcast_buffers": null,
+  "dataloader_pin_memory": true,
+  "dataloader_persistent_workers": false,
+  "skip_memory_metrics": true,
+  "use_legacy_prediction_loop": false,
+  "push_to_hub": false,
+  "resume_from_checkpoint": null,
+  "hub_model_id": null,
+  "hub_strategy": "every_save",
+  "hub_token": null,
+  "hub_private_repo": null,
+  "hub_always_push": false,
+  "gradient_checkpointing": true,
+  "gradient_checkpointing_kwargs": "{\"use_reentrant\": false}",
+  "include_inputs_for_metrics": false,
+  "include_for_metrics": [],
+  "eval_do_concat_batches": true,
+  "fp16_backend": "auto",
+  "evaluation_strategy": "steps",
+  "push_to_hub_model_id": null,
+  "push_to_hub_organization": null,
+  "push_to_hub_token": null,
+  "mp_parameters": "",
+  "auto_find_batch_size": false,
+  "full_determinism": false,
+  "torchdynamo": null,
+  "ray_scope": "last",
+  "ddp_timeout": 18000000,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null,
+  "dispatch_batches": null,
+  "split_batches": null,
+  "include_tokens_per_second": false,
+  "include_num_input_tokens_seen": false,
+  "neftune_noise_alpha": null,
+  "optim_target_modules": null,
+  "batch_eval_metrics": false,
+  "eval_on_start": false,
+  "use_liger_kernel": false,
+  "eval_use_gather_object": false,
+  "average_tokens_across_devices": false,
+  "sortish_sampler": false,
+  "predict_with_generate": false,
+  "generation_max_length": null,
+  "generation_num_beams": null,
+  "generation_config": null,
+  "tuner_backend": "peft",
+  "vit_gradient_checkpointing": null,
+  "router_aux_loss_coef": 0.0,
+  "enable_dft_loss": false,
+  "enable_channel_loss": false,
+  "check_model": true,
+  "acc_strategy": "token",
+  "train_dataloader_shuffle": true,
+  "max_epochs": null,
+  "aligner_lr": 1e-05,
+  "vit_lr": 5e-06,
+  "use_logits_to_keep": null,
+  "ds3_gather_for_generation": true,
+  "resume_only_model": false,
+  "optimizer": null,
+  "loss_type": null,
+  "metric": null,
+  "eval_use_evalscope": false,
+  "eval_dataset": [],
+  "eval_dataset_args": null,
+  "eval_limit": null,
+  "eval_generation_config": null,
+  "extra_eval_args": null,
+  "use_flash_ckpt": false,
+  "use_ray": false,
+  "ray_exp_name": null,
+  "device_groups": null,
+  "model": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "model_type": "step_audio2_mini",
+  "model_revision": null,
+  "task_type": "causal_lm",
+  "torch_dtype": "bfloat16",
+  "attn_impl": null,
+  "new_special_tokens": [],
+  "num_labels": null,
+  "problem_type": null,
+  "rope_scaling": null,
+  "device_map": null,
+  "max_memory": {},
+  "max_model_len": null,
+  "local_repo_path": null,
+  "init_strategy": null,
+  "template": "step_audio2_mini",
+  "system": null,
+  "max_length": 16384,
+  "truncation_strategy": "delete",
+  "max_pixels": null,
+  "agent_template": null,
+  "norm_bbox": null,
+  "use_chat_template": true,
+  "padding_free": false,
+  "padding_side": "right",
+  "loss_scale": "default",
+  "sequence_parallel_size": 1,
+  "response_prefix": null,
+  "template_backend": "swift",
+  "dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_train90.jsonl"
+  ],
+  "val_dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "cached_dataset": [],
+  "cached_val_dataset": [],
+  "split_dataset_ratio": 0.0,
+  "dataset_num_proc": 1,
+  "load_from_cache_file": false,
+  "dataset_shuffle": true,
+  "val_dataset_shuffle": false,
+  "streaming": false,
+  "interleave_prob": null,
+  "stopping_strategy": "first_exhausted",
+  "shuffle_buffer_size": 1000,
+  "download_mode": "reuse_dataset_if_exists",
+  "columns": {},
+  "strict": false,
+  "model_name": null,
+  "model_author": null,
+  "custom_dataset_info": [],
+  "quant_method": null,
+  "quant_bits": null,
+  "hqq_axis": null,
+  "bnb_4bit_compute_dtype": "bfloat16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "max_new_tokens": 64,
+  "temperature": 0.0,
+  "top_k": null,
+  "top_p": null,
+  "repetition_penalty": null,
+  "num_beams": 1,
+  "stream": false,
+  "stop_words": [],
+  "logprobs": false,
+  "top_logprobs": null,
+  "ckpt_dir": null,
+  "lora_modules": [],
+  "train_type": "lora",
+  "adapters": [],
+  "external_plugins": [],
+  "model_kwargs": {},
+  "load_args": false,
+  "load_data_args": false,
+  "packing": false,
+  "packing_length": null,
+  "packing_num_proc": 1,
+  "lazy_tokenize": true,
+  "custom_register_path": [],
+  "use_hf": false,
+  "ignore_args_error": false,
+  "use_swift_lora": false,
+  "freeze_parameters": [],
+  "freeze_parameters_regex": null,
+  "freeze_parameters_ratio": 0.0,
+  "trainable_parameters": [],
+  "trainable_parameters_regex": null,
+  "freeze_llm": true,
+  "freeze_vit": false,
+  "freeze_aligner": false,
+  "target_modules": [
+    "all-linear"
+  ],
+  "target_regex": null,
+  "target_parameters": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "lora_bias": "none",
+  "lora_dtype": null,
+  "lorap_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "lora_ga_batch_size": 2,
+  "lora_ga_iters": 2,
+  "lora_ga_max_length": 1024,
+  "lora_ga_direction": "ArB2r",
+  "lora_ga_scale": "stable",
+  "lora_ga_stable_gamma": 16,
+  "init_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "use_galore": false,
+  "galore_target_modules": null,
+  "galore_rank": 128,
+  "galore_update_proj_gap": 50,
+  "galore_scale": 1.0,
+  "galore_proj_type": "std",
+  "galore_optim_per_parameter": false,
+  "galore_with_embedding": false,
+  "galore_quantization": false,
+  "galore_proj_quant": false,
+  "galore_proj_bits": 4,
+  "galore_proj_group_size": 256,
+  "galore_cos_threshold": 0.4,
+  "galore_gamma_proj": 2,
+  "galore_queue_size": 5,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "lisa_activated_layers": 0,
+  "lisa_step_interval": 20,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "swanlab_token": null,
+  "swanlab_project": null,
+  "swanlab_workspace": null,
+  "swanlab_exp_name": null,
+  "swanlab_lark_webhook_url": null,
+  "swanlab_lark_secret": null,
+  "swanlab_mode": "cloud",
+  "add_version": true,
+  "create_checkpoint_symlink": false,
+  "zero_hpz_partition_size": null,
+  "deepspeed_autotp_size": null,
+  "early_stop_interval": null,
+  "rank": -1,
+  "global_world_size": 1,
+  "local_world_size": 1,
+  "model_suffix": "Step-Audio-2-mini",
+  "model_info": "ModelInfo(model_type='step_audio2_mini', model_dir='/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini', torch_dtype=torch.bfloat16, max_model_len=16384, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=True, config=None, task_type='causal_lm', num_labels=None)",
+  "model_meta": "ModelMeta(model_type='step_audio2_mini', model_groups=[ModelGroup(models=[Model(ms_model_id='stepfun-ai/Step-Audio-2-mini', hf_model_id='stepfun-ai/Step-Audio-2-mini', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='step_audio2_mini', get_function=<function get_model_tokenizer_step_audio2_mini at 0x7f32c017c4c0>, model_arch=MultiModelKeys(arch_name='step_audio2_mini', embedding=None, module_list=None, lm_head=None, q_proj=None, k_proj=None, v_proj=None, o_proj=None, attention=None, mlp=None, down_proj=None, qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None, language_model=['model', 'lm_head'], aligner=['adapter'], vision_tower=['encoder'], generator=[]), architectures=['StepAudio2ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=True, is_reward=False, is_reranker=False, task_type=None, ignore_patterns=None, requires=['transformers==4.53.3', 'torchaudio', 'librosa'], tags=['audio'])",
+  "model_dir": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "_val_dataset_exists": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "hub": "<class 'swift.hub.hub.MSHub'>",
+  "training_args": "Seq2SeqTrainingArguments(output_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=16, per_device_eval_batch_size=16, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=4, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=5e-06, weight_decay=0.01, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=1.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.02, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=100, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=2000, save_total_limit=40, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=2000, dataloader_num_workers=1, dataloader_prefetch_factor=10, past_index=-1, run_name='step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['wandb'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs={'use_reentrant': False}, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy='steps', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, eval_use_gather_object=False, average_tokens_across_devices=None, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=1e-05, vit_lr=5e-06, use_logits_to_keep=None, ds3_gather_for_generation=True, resume_only_model=False, optimizer='multimodal', loss_type=None, metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='lora', local_repo_path=None, galore_config=None, padding_side='right', padding_free=False, task_type='causal_lm', problem_type=None)"
+}

v0-20260123-014416/checkpoint-2000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2efe81479a806272b0f65c224098b8615b114494e4739155437b68403abf17f9
+size 24205999

v0-20260123-014416/checkpoint-2000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba831bcb8faf9f5101c94e9dfbf8557a87bf40c6ea33f78a837bdfcb9953b0a3
+size 14645

v0-20260123-014416/checkpoint-2000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24cc5c369df9f5313e40bf46692e1978487bd2ab62aa6773313c29be84334c1b
+size 1465

v0-20260123-014416/checkpoint-2000/trainer_state.json

@@ -0,0 +1,210 @@
+{
+  "best_metric": 0.55640507,
+  "best_model_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-2000",
+  "epoch": 0.21372088053002777,
+  "eval_steps": 2000,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00010686044026501389,
+      "grad_norm": 1.6953125,
+      "learning_rate": 2.6595744680851065e-08,
+      "loss": 0.7167500853538513,
+      "step": 1,
+      "token_acc": 0.8417487684729064
+    },
+    {
+      "epoch": 0.010686044026501388,
+      "grad_norm": 1.4453125,
+      "learning_rate": 2.6595744680851065e-06,
+      "loss": 0.6973034637142913,
+      "step": 100,
+      "token_acc": 0.8467176205711577
+    },
+    {
+      "epoch": 0.021372088053002777,
+      "grad_norm": 1.046875,
+      "learning_rate": 4.999978873236658e-06,
+      "loss": 0.6577812957763672,
+      "step": 200,
+      "token_acc": 0.8531807124137688
+    },
+    {
+      "epoch": 0.03205813207950417,
+      "grad_norm": 1.328125,
+      "learning_rate": 4.998159847367393e-06,
+      "loss": 0.6022152709960937,
+      "step": 300,
+      "token_acc": 0.8620946875218537
+    },
+    {
+      "epoch": 0.04274417610600555,
+      "grad_norm": 1.2421875,
+      "learning_rate": 4.993408991262632e-06,
+      "loss": 0.5575482177734375,
+      "step": 400,
+      "token_acc": 0.8680139909957781
+    },
+    {
+      "epoch": 0.05343022013250694,
+      "grad_norm": 1.8984375,
+      "learning_rate": 4.985731880507775e-06,
+      "loss": 0.5264178466796875,
+      "step": 500,
+      "token_acc": 0.8696029378557931
+    },
+    {
+      "epoch": 0.06411626415900834,
+      "grad_norm": 1.1171875,
+      "learning_rate": 4.975137524928774e-06,
+      "loss": 0.4771532440185547,
+      "step": 600,
+      "token_acc": 0.8748165713816137
+    },
+    {
+      "epoch": 0.07480230818550973,
+      "grad_norm": 1.4765625,
+      "learning_rate": 4.961638358018247e-06,
+      "loss": 0.4643104553222656,
+      "step": 700,
+      "token_acc": 0.8754063671076119
+    },
+    {
+      "epoch": 0.0854883522120111,
+      "grad_norm": 1.109375,
+      "learning_rate": 4.945250222343571e-06,
+      "loss": 0.4561585235595703,
+      "step": 800,
+      "token_acc": 0.8795579017210171
+    },
+    {
+      "epoch": 0.0961743962385125,
+      "grad_norm": 1.171875,
+      "learning_rate": 4.925992350954112e-06,
+      "loss": 0.4534299850463867,
+      "step": 900,
+      "token_acc": 0.8851879828938838
+    },
+    {
+      "epoch": 0.10686044026501389,
+      "grad_norm": 1.171875,
+      "learning_rate": 4.903887344809403e-06,
+      "loss": 0.42084632873535155,
+      "step": 1000,
+      "token_acc": 0.8909899302321495
+    },
+    {
+      "epoch": 0.11754648429151528,
+      "grad_norm": 1.390625,
+      "learning_rate": 4.8789611462547414e-06,
+      "loss": 0.43503849029541014,
+      "step": 1100,
+      "token_acc": 0.8900309663514769
+    },
+    {
+      "epoch": 0.12823252831801668,
+      "grad_norm": 1.21875,
+      "learning_rate": 4.851243008575364e-06,
+      "loss": 0.4078071594238281,
+      "step": 1200,
+      "token_acc": 0.8972086577408744
+    },
+    {
+      "epoch": 0.13891857234451807,
+      "grad_norm": 1.3046875,
+      "learning_rate": 4.8207654616649035e-06,
+      "loss": 0.4169163131713867,
+      "step": 1300,
+      "token_acc": 0.8944712980277317
+    },
+    {
+      "epoch": 0.14960461637101946,
+      "grad_norm": 1.25,
+      "learning_rate": 4.7875642738484475e-06,
+      "loss": 0.41801651000976564,
+      "step": 1400,
+      "token_acc": 0.8954553844138678
+    },
+    {
+      "epoch": 0.16029066039752085,
+      "grad_norm": 1.0390625,
+      "learning_rate": 4.7516784099049705e-06,
+      "loss": 0.42521297454833984,
+      "step": 1500,
+      "token_acc": 0.8927475600806661
+    },
+    {
+      "epoch": 0.1709767044240222,
+      "grad_norm": 1.0625,
+      "learning_rate": 4.713149985338428e-06,
+      "loss": 0.40242053985595705,
+      "step": 1600,
+      "token_acc": 0.8985869393070196
+    },
+    {
+      "epoch": 0.1816627484505236,
+      "grad_norm": 1.6953125,
+      "learning_rate": 4.672024216951176e-06,
+      "loss": 0.39753116607666017,
+      "step": 1700,
+      "token_acc": 0.9003083728795708
+    },
+    {
+      "epoch": 0.192348792477025,
+      "grad_norm": 1.328125,
+      "learning_rate": 4.628349369777712e-06,
+      "loss": 0.40670780181884764,
+      "step": 1800,
+      "token_acc": 0.8985113659223496
+    },
+    {
+      "epoch": 0.20303483650352638,
+      "grad_norm": 1.609375,
+      "learning_rate": 4.58217670044103e-06,
+      "loss": 0.4040363311767578,
+      "step": 1900,
+      "token_acc": 0.898509424415663
+    },
+    {
+      "epoch": 0.21372088053002777,
+      "grad_norm": 1.3203125,
+      "learning_rate": 4.533560396998055e-06,
+      "loss": 0.3897912216186523,
+      "step": 2000,
+      "token_acc": 0.9022499382853919
+    },
+    {
+      "epoch": 0.21372088053002777,
+      "eval_loss": 0.5564050674438477,
+      "eval_runtime": 1628.334,
+      "eval_samples_per_second": 20.433,
+      "eval_steps_per_second": 1.277,
+      "eval_token_acc": 0.8750522224756502,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 9358,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 2000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.607272128827302e+18,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

v0-20260123-014416/checkpoint-2000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e61a5ccecfe4850e11b9fb1b8bea0737456f069a42a114f94ef68317432e4e8
+size 7185

v0-20260123-014416/checkpoint-4000/README.md

@@ -0,0 +1,202 @@
+---
+base_model: /data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.11.1

v0-20260123-014416/checkpoint-4000/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [],
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "^(encoder.*\\.(query|mlp.0|mlp.2|out|value|key)|adapter.*\\.(linear1|linear2))$",
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

v0-20260123-014416/checkpoint-4000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f315dcc36f8a2d56b54570ce77e301a8a30c57d4fd0824ce5922f44bab8ee778
+size 11988912

v0-20260123-014416/checkpoint-4000/additional_config.json

@@ -0,0 +1 @@
+{"lora_dtype": null, "lorap_lr_ratio": null, "lorap_emb_lr": 1e-06}

v0-20260123-014416/checkpoint-4000/args.json

@@ -0,0 +1,351 @@
+{
+  "output_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416",
+  "overwrite_output_dir": false,
+  "do_train": false,
+  "do_eval": false,
+  "do_predict": false,
+  "eval_strategy": "steps",
+  "prediction_loss_only": false,
+  "per_device_train_batch_size": 16,
+  "per_device_eval_batch_size": 16,
+  "per_gpu_train_batch_size": null,
+  "per_gpu_eval_batch_size": null,
+  "gradient_accumulation_steps": 4,
+  "eval_accumulation_steps": null,
+  "eval_delay": 0,
+  "torch_empty_cache_steps": null,
+  "learning_rate": 5e-06,
+  "weight_decay": 0.01,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "max_grad_norm": 1.0,
+  "num_train_epochs": 1.0,
+  "max_steps": -1,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": null,
+  "warmup_ratio": 0.02,
+  "warmup_steps": 0,
+  "log_level": "passive",
+  "log_level_replica": "warning",
+  "log_on_each_node": true,
+  "logging_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/runs",
+  "logging_strategy": "steps",
+  "logging_first_step": true,
+  "logging_steps": 100,
+  "logging_nan_inf_filter": true,
+  "save_strategy": "steps",
+  "save_steps": 2000.0,
+  "save_total_limit": 40,
+  "save_safetensors": true,
+  "save_on_each_node": false,
+  "save_only_model": false,
+  "restore_callback_states_from_checkpoint": false,
+  "no_cuda": false,
+  "use_cpu": false,
+  "use_mps_device": false,
+  "seed": 42,
+  "data_seed": 42,
+  "jit_mode_eval": false,
+  "use_ipex": false,
+  "bf16": true,
+  "fp16": false,
+  "fp16_opt_level": "O1",
+  "half_precision_backend": "auto",
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "local_rank": -1,
+  "ddp_backend": null,
+  "tpu_num_cores": null,
+  "tpu_metrics_debug": false,
+  "debug": null,
+  "dataloader_drop_last": false,
+  "eval_steps": 2000.0,
+  "dataloader_num_workers": null,
+  "dataloader_prefetch_factor": null,
+  "past_index": -1,
+  "run_name": "step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF",
+  "disable_tqdm": null,
+  "remove_unused_columns": true,
+  "label_names": null,
+  "load_best_model_at_end": false,
+  "metric_for_best_model": "loss",
+  "greater_is_better": false,
+  "ignore_data_skip": false,
+  "fsdp": "",
+  "fsdp_min_num_params": 0,
+  "fsdp_config": null,
+  "fsdp_transformer_layer_cls_to_wrap": null,
+  "accelerator_config": {
+    "dispatch_batches": false
+  },
+  "deepspeed": null,
+  "label_smoothing_factor": 0.0,
+  "optim": "adamw_torch",
+  "optim_args": null,
+  "adafactor": false,
+  "group_by_length": false,
+  "length_column_name": "length",
+  "report_to": [
+    "wandb"
+  ],
+  "ddp_find_unused_parameters": null,
+  "ddp_bucket_cap_mb": null,
+  "ddp_broadcast_buffers": null,
+  "dataloader_pin_memory": true,
+  "dataloader_persistent_workers": false,
+  "skip_memory_metrics": true,
+  "use_legacy_prediction_loop": false,
+  "push_to_hub": false,
+  "resume_from_checkpoint": null,
+  "hub_model_id": null,
+  "hub_strategy": "every_save",
+  "hub_token": null,
+  "hub_private_repo": null,
+  "hub_always_push": false,
+  "gradient_checkpointing": true,
+  "gradient_checkpointing_kwargs": "{\"use_reentrant\": false}",
+  "include_inputs_for_metrics": false,
+  "include_for_metrics": [],
+  "eval_do_concat_batches": true,
+  "fp16_backend": "auto",
+  "evaluation_strategy": "steps",
+  "push_to_hub_model_id": null,
+  "push_to_hub_organization": null,
+  "push_to_hub_token": null,
+  "mp_parameters": "",
+  "auto_find_batch_size": false,
+  "full_determinism": false,
+  "torchdynamo": null,
+  "ray_scope": "last",
+  "ddp_timeout": 18000000,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null,
+  "dispatch_batches": null,
+  "split_batches": null,
+  "include_tokens_per_second": false,
+  "include_num_input_tokens_seen": false,
+  "neftune_noise_alpha": null,
+  "optim_target_modules": null,
+  "batch_eval_metrics": false,
+  "eval_on_start": false,
+  "use_liger_kernel": false,
+  "eval_use_gather_object": false,
+  "average_tokens_across_devices": false,
+  "sortish_sampler": false,
+  "predict_with_generate": false,
+  "generation_max_length": null,
+  "generation_num_beams": null,
+  "generation_config": null,
+  "tuner_backend": "peft",
+  "vit_gradient_checkpointing": null,
+  "router_aux_loss_coef": 0.0,
+  "enable_dft_loss": false,
+  "enable_channel_loss": false,
+  "check_model": true,
+  "acc_strategy": "token",
+  "train_dataloader_shuffle": true,
+  "max_epochs": null,
+  "aligner_lr": 1e-05,
+  "vit_lr": 5e-06,
+  "use_logits_to_keep": null,
+  "ds3_gather_for_generation": true,
+  "resume_only_model": false,
+  "optimizer": null,
+  "loss_type": null,
+  "metric": null,
+  "eval_use_evalscope": false,
+  "eval_dataset": [],
+  "eval_dataset_args": null,
+  "eval_limit": null,
+  "eval_generation_config": null,
+  "extra_eval_args": null,
+  "use_flash_ckpt": false,
+  "use_ray": false,
+  "ray_exp_name": null,
+  "device_groups": null,
+  "model": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "model_type": "step_audio2_mini",
+  "model_revision": null,
+  "task_type": "causal_lm",
+  "torch_dtype": "bfloat16",
+  "attn_impl": null,
+  "new_special_tokens": [],
+  "num_labels": null,
+  "problem_type": null,
+  "rope_scaling": null,
+  "device_map": null,
+  "max_memory": {},
+  "max_model_len": null,
+  "local_repo_path": null,
+  "init_strategy": null,
+  "template": "step_audio2_mini",
+  "system": null,
+  "max_length": 16384,
+  "truncation_strategy": "delete",
+  "max_pixels": null,
+  "agent_template": null,
+  "norm_bbox": null,
+  "use_chat_template": true,
+  "padding_free": false,
+  "padding_side": "right",
+  "loss_scale": "default",
+  "sequence_parallel_size": 1,
+  "response_prefix": null,
+  "template_backend": "swift",
+  "dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_train90.jsonl"
+  ],
+  "val_dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "cached_dataset": [],
+  "cached_val_dataset": [],
+  "split_dataset_ratio": 0.0,
+  "dataset_num_proc": 1,
+  "load_from_cache_file": false,
+  "dataset_shuffle": true,
+  "val_dataset_shuffle": false,
+  "streaming": false,
+  "interleave_prob": null,
+  "stopping_strategy": "first_exhausted",
+  "shuffle_buffer_size": 1000,
+  "download_mode": "reuse_dataset_if_exists",
+  "columns": {},
+  "strict": false,
+  "model_name": null,
+  "model_author": null,
+  "custom_dataset_info": [],
+  "quant_method": null,
+  "quant_bits": null,
+  "hqq_axis": null,
+  "bnb_4bit_compute_dtype": "bfloat16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "max_new_tokens": 64,
+  "temperature": 0.0,
+  "top_k": null,
+  "top_p": null,
+  "repetition_penalty": null,
+  "num_beams": 1,
+  "stream": false,
+  "stop_words": [],
+  "logprobs": false,
+  "top_logprobs": null,
+  "ckpt_dir": null,
+  "lora_modules": [],
+  "train_type": "lora",
+  "adapters": [],
+  "external_plugins": [],
+  "model_kwargs": {},
+  "load_args": false,
+  "load_data_args": false,
+  "packing": false,
+  "packing_length": null,
+  "packing_num_proc": 1,
+  "lazy_tokenize": true,
+  "custom_register_path": [],
+  "use_hf": false,
+  "ignore_args_error": false,
+  "use_swift_lora": false,
+  "freeze_parameters": [],
+  "freeze_parameters_regex": null,
+  "freeze_parameters_ratio": 0.0,
+  "trainable_parameters": [],
+  "trainable_parameters_regex": null,
+  "freeze_llm": true,
+  "freeze_vit": false,
+  "freeze_aligner": false,
+  "target_modules": [
+    "all-linear"
+  ],
+  "target_regex": null,
+  "target_parameters": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "lora_bias": "none",
+  "lora_dtype": null,
+  "lorap_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "lora_ga_batch_size": 2,
+  "lora_ga_iters": 2,
+  "lora_ga_max_length": 1024,
+  "lora_ga_direction": "ArB2r",
+  "lora_ga_scale": "stable",
+  "lora_ga_stable_gamma": 16,
+  "init_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "use_galore": false,
+  "galore_target_modules": null,
+  "galore_rank": 128,
+  "galore_update_proj_gap": 50,
+  "galore_scale": 1.0,
+  "galore_proj_type": "std",
+  "galore_optim_per_parameter": false,
+  "galore_with_embedding": false,
+  "galore_quantization": false,
+  "galore_proj_quant": false,
+  "galore_proj_bits": 4,
+  "galore_proj_group_size": 256,
+  "galore_cos_threshold": 0.4,
+  "galore_gamma_proj": 2,
+  "galore_queue_size": 5,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "lisa_activated_layers": 0,
+  "lisa_step_interval": 20,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "swanlab_token": null,
+  "swanlab_project": null,
+  "swanlab_workspace": null,
+  "swanlab_exp_name": null,
+  "swanlab_lark_webhook_url": null,
+  "swanlab_lark_secret": null,
+  "swanlab_mode": "cloud",
+  "add_version": true,
+  "create_checkpoint_symlink": false,
+  "zero_hpz_partition_size": null,
+  "deepspeed_autotp_size": null,
+  "early_stop_interval": null,
+  "rank": -1,
+  "global_world_size": 1,
+  "local_world_size": 1,
+  "model_suffix": "Step-Audio-2-mini",
+  "model_info": "ModelInfo(model_type='step_audio2_mini', model_dir='/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini', torch_dtype=torch.bfloat16, max_model_len=16384, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=True, config=None, task_type='causal_lm', num_labels=None)",
+  "model_meta": "ModelMeta(model_type='step_audio2_mini', model_groups=[ModelGroup(models=[Model(ms_model_id='stepfun-ai/Step-Audio-2-mini', hf_model_id='stepfun-ai/Step-Audio-2-mini', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='step_audio2_mini', get_function=<function get_model_tokenizer_step_audio2_mini at 0x7f32c017c4c0>, model_arch=MultiModelKeys(arch_name='step_audio2_mini', embedding=None, module_list=None, lm_head=None, q_proj=None, k_proj=None, v_proj=None, o_proj=None, attention=None, mlp=None, down_proj=None, qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None, language_model=['model', 'lm_head'], aligner=['adapter'], vision_tower=['encoder'], generator=[]), architectures=['StepAudio2ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=True, is_reward=False, is_reranker=False, task_type=None, ignore_patterns=None, requires=['transformers==4.53.3', 'torchaudio', 'librosa'], tags=['audio'])",
+  "model_dir": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "_val_dataset_exists": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "hub": "<class 'swift.hub.hub.MSHub'>",
+  "training_args": "Seq2SeqTrainingArguments(output_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=16, per_device_eval_batch_size=16, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=4, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=5e-06, weight_decay=0.01, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=1.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.02, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=100, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=2000, save_total_limit=40, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=2000, dataloader_num_workers=1, dataloader_prefetch_factor=10, past_index=-1, run_name='step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['wandb'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs={'use_reentrant': False}, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy='steps', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, eval_use_gather_object=False, average_tokens_across_devices=None, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=1e-05, vit_lr=5e-06, use_logits_to_keep=None, ds3_gather_for_generation=True, resume_only_model=False, optimizer='multimodal', loss_type=None, metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='lora', local_repo_path=None, galore_config=None, padding_side='right', padding_free=False, task_type='causal_lm', problem_type=None)"
+}

v0-20260123-014416/checkpoint-4000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b3ddf443d798031411e6f911b39d2e0b74db8a9e149794dafbaa01adc706b09
+size 24205999

v0-20260123-014416/checkpoint-4000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65a0256738d0a2b4fb93409e385a10b55cf44ea01734625f77b31f99a543f3a1
+size 14645

v0-20260123-014416/checkpoint-4000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5463e23ca1127ec48142e74ff51708175807ecfcea702865fb9c8596fa6ce8c0
+size 1465

v0-20260123-014416/checkpoint-4000/trainer_state.json

@@ -0,0 +1,379 @@
+{
+  "best_metric": 0.55598271,
+  "best_model_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000",
+  "epoch": 0.42744176106005555,
+  "eval_steps": 2000,
+  "global_step": 4000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00010686044026501389,
+      "grad_norm": 1.6953125,
+      "learning_rate": 2.6595744680851065e-08,
+      "loss": 0.7167500853538513,
+      "step": 1,
+      "token_acc": 0.8417487684729064
+    },
+    {
+      "epoch": 0.010686044026501388,
+      "grad_norm": 1.4453125,
+      "learning_rate": 2.6595744680851065e-06,
+      "loss": 0.6973034637142913,
+      "step": 100,
+      "token_acc": 0.8467176205711577
+    },
+    {
+      "epoch": 0.021372088053002777,
+      "grad_norm": 1.046875,
+      "learning_rate": 4.999978873236658e-06,
+      "loss": 0.6577812957763672,
+      "step": 200,
+      "token_acc": 0.8531807124137688
+    },
+    {
+      "epoch": 0.03205813207950417,
+      "grad_norm": 1.328125,
+      "learning_rate": 4.998159847367393e-06,
+      "loss": 0.6022152709960937,
+      "step": 300,
+      "token_acc": 0.8620946875218537
+    },
+    {
+      "epoch": 0.04274417610600555,
+      "grad_norm": 1.2421875,
+      "learning_rate": 4.993408991262632e-06,
+      "loss": 0.5575482177734375,
+      "step": 400,
+      "token_acc": 0.8680139909957781
+    },
+    {
+      "epoch": 0.05343022013250694,
+      "grad_norm": 1.8984375,
+      "learning_rate": 4.985731880507775e-06,
+      "loss": 0.5264178466796875,
+      "step": 500,
+      "token_acc": 0.8696029378557931
+    },
+    {
+      "epoch": 0.06411626415900834,
+      "grad_norm": 1.1171875,
+      "learning_rate": 4.975137524928774e-06,
+      "loss": 0.4771532440185547,
+      "step": 600,
+      "token_acc": 0.8748165713816137
+    },
+    {
+      "epoch": 0.07480230818550973,
+      "grad_norm": 1.4765625,
+      "learning_rate": 4.961638358018247e-06,
+      "loss": 0.4643104553222656,
+      "step": 700,
+      "token_acc": 0.8754063671076119
+    },
+    {
+      "epoch": 0.0854883522120111,
+      "grad_norm": 1.109375,
+      "learning_rate": 4.945250222343571e-06,
+      "loss": 0.4561585235595703,
+      "step": 800,
+      "token_acc": 0.8795579017210171
+    },
+    {
+      "epoch": 0.0961743962385125,
+      "grad_norm": 1.171875,
+      "learning_rate": 4.925992350954112e-06,
+      "loss": 0.4534299850463867,
+      "step": 900,
+      "token_acc": 0.8851879828938838
+    },
+    {
+      "epoch": 0.10686044026501389,
+      "grad_norm": 1.171875,
+      "learning_rate": 4.903887344809403e-06,
+      "loss": 0.42084632873535155,
+      "step": 1000,
+      "token_acc": 0.8909899302321495
+    },
+    {
+      "epoch": 0.11754648429151528,
+      "grad_norm": 1.390625,
+      "learning_rate": 4.8789611462547414e-06,
+      "loss": 0.43503849029541014,
+      "step": 1100,
+      "token_acc": 0.8900309663514769
+    },
+    {
+      "epoch": 0.12823252831801668,
+      "grad_norm": 1.21875,
+      "learning_rate": 4.851243008575364e-06,
+      "loss": 0.4078071594238281,
+      "step": 1200,
+      "token_acc": 0.8972086577408744
+    },
+    {
+      "epoch": 0.13891857234451807,
+      "grad_norm": 1.3046875,
+      "learning_rate": 4.8207654616649035e-06,
+      "loss": 0.4169163131713867,
+      "step": 1300,
+      "token_acc": 0.8944712980277317
+    },
+    {
+      "epoch": 0.14960461637101946,
+      "grad_norm": 1.25,
+      "learning_rate": 4.7875642738484475e-06,
+      "loss": 0.41801651000976564,
+      "step": 1400,
+      "token_acc": 0.8954553844138678
+    },
+    {
+      "epoch": 0.16029066039752085,
+      "grad_norm": 1.0390625,
+      "learning_rate": 4.7516784099049705e-06,
+      "loss": 0.42521297454833984,
+      "step": 1500,
+      "token_acc": 0.8927475600806661
+    },
+    {
+      "epoch": 0.1709767044240222,
+      "grad_norm": 1.0625,
+      "learning_rate": 4.713149985338428e-06,
+      "loss": 0.40242053985595705,
+      "step": 1600,
+      "token_acc": 0.8985869393070196
+    },
+    {
+      "epoch": 0.1816627484505236,
+      "grad_norm": 1.6953125,
+      "learning_rate": 4.672024216951176e-06,
+      "loss": 0.39753116607666017,
+      "step": 1700,
+      "token_acc": 0.9003083728795708
+    },
+    {
+      "epoch": 0.192348792477025,
+      "grad_norm": 1.328125,
+      "learning_rate": 4.628349369777712e-06,
+      "loss": 0.40670780181884764,
+      "step": 1800,
+      "token_acc": 0.8985113659223496
+    },
+    {
+      "epoch": 0.20303483650352638,
+      "grad_norm": 1.609375,
+      "learning_rate": 4.58217670044103e-06,
+      "loss": 0.4040363311767578,
+      "step": 1900,
+      "token_acc": 0.898509424415663
+    },
+    {
+      "epoch": 0.21372088053002777,
+      "grad_norm": 1.3203125,
+      "learning_rate": 4.533560396998055e-06,
+      "loss": 0.3897912216186523,
+      "step": 2000,
+      "token_acc": 0.9022499382853919
+    },
+    {
+      "epoch": 0.21372088053002777,
+      "eval_loss": 0.5564050674438477,
+      "eval_runtime": 1628.334,
+      "eval_samples_per_second": 20.433,
+      "eval_steps_per_second": 1.277,
+      "eval_token_acc": 0.8750522224756502,
+      "step": 2000
+    },
+    {
+      "epoch": 0.22440692455652916,
+      "grad_norm": 1.390625,
+      "learning_rate": 4.482557515344757e-06,
+      "loss": 0.39870925903320314,
+      "step": 2100,
+      "token_acc": 0.8990360044410765
+    },
+    {
+      "epoch": 0.23509296858303055,
+      "grad_norm": 1.140625,
+      "learning_rate": 4.429227912255589e-06,
+      "loss": 0.38395835876464846,
+      "step": 2200,
+      "token_acc": 0.902773223403757
+    },
+    {
+      "epoch": 0.24577901260953194,
+      "grad_norm": 0.96484375,
+      "learning_rate": 4.3736341751358166e-06,
+      "loss": 0.39471343994140623,
+      "step": 2300,
+      "token_acc": 0.9010956332106893
+    },
+    {
+      "epoch": 0.25646505663603336,
+      "grad_norm": 1.4140625,
+      "learning_rate": 4.315841548569193e-06,
+      "loss": 0.39796932220458986,
+      "step": 2400,
+      "token_acc": 0.9007656570069128
+    },
+    {
+      "epoch": 0.2671511006625347,
+      "grad_norm": 1.265625,
+      "learning_rate": 4.255917857747183e-06,
+      "loss": 0.3913582992553711,
+      "step": 2500,
+      "token_acc": 0.9015067434468572
+    },
+    {
+      "epoch": 0.27783714468903614,
+      "grad_norm": 1.46875,
+      "learning_rate": 4.193933428869592e-06,
+      "loss": 0.38879886627197263,
+      "step": 2600,
+      "token_acc": 0.9019667027692453
+    },
+    {
+      "epoch": 0.2885231887155375,
+      "grad_norm": 2.4375,
+      "learning_rate": 4.129961006610028e-06,
+      "loss": 0.3972347259521484,
+      "step": 2700,
+      "token_acc": 0.9001422292689183
+    },
+    {
+      "epoch": 0.2992092327420389,
+      "grad_norm": 1.2578125,
+      "learning_rate": 4.064075668743043e-06,
+      "loss": 0.3885000991821289,
+      "step": 2800,
+      "token_acc": 0.9027788505528588
+    },
+    {
+      "epoch": 0.3098952767685403,
+      "grad_norm": 1.265625,
+      "learning_rate": 3.996354738033166e-06,
+      "loss": 0.39520988464355467,
+      "step": 2900,
+      "token_acc": 0.9011279097672186
+    },
+    {
+      "epoch": 0.3205813207950417,
+      "grad_norm": 0.921875,
+      "learning_rate": 3.926877691489222e-06,
+      "loss": 0.3871217346191406,
+      "step": 3000,
+      "token_acc": 0.9022083313787118
+    },
+    {
+      "epoch": 0.33126736482154306,
+      "grad_norm": 1.375,
+      "learning_rate": 3.855726067090433e-06,
+      "loss": 0.3810308074951172,
+      "step": 3100,
+      "token_acc": 0.9034641205430405
+    },
+    {
+      "epoch": 0.3419534088480444,
+      "grad_norm": 1.5546875,
+      "learning_rate": 3.7829833680937804e-06,
+      "loss": 0.37891117095947263,
+      "step": 3200,
+      "token_acc": 0.9025715320536037
+    },
+    {
+      "epoch": 0.35263945287454584,
+      "grad_norm": 1.59375,
+      "learning_rate": 3.7087349650349234e-06,
+      "loss": 0.40724864959716794,
+      "step": 3300,
+      "token_acc": 0.8990014549892049
+    },
+    {
+      "epoch": 0.3633254969010472,
+      "grad_norm": 1.1171875,
+      "learning_rate": 3.6330679955376843e-06,
+      "loss": 0.39273796081542967,
+      "step": 3400,
+      "token_acc": 0.900933421493117
+    },
+    {
+      "epoch": 0.3740115409275486,
+      "grad_norm": 1.5,
+      "learning_rate": 3.556071262049684e-06,
+      "loss": 0.3975730514526367,
+      "step": 3500,
+      "token_acc": 0.8998019157822584
+    },
+    {
+      "epoch": 0.38469758495405,
+      "grad_norm": 2.140625,
+      "learning_rate": 3.477835127624158e-06,
+      "loss": 0.37645999908447264,
+      "step": 3600,
+      "token_acc": 0.9051004274755519
+    },
+    {
+      "epoch": 0.3953836289805514,
+      "grad_norm": 1.1796875,
+      "learning_rate": 3.39845140987024e-06,
+      "loss": 0.4185586166381836,
+      "step": 3700,
+      "token_acc": 0.8963374986700084
+    },
+    {
+      "epoch": 0.40606967300705277,
+      "grad_norm": 1.0390625,
+      "learning_rate": 3.318013273196187e-06,
+      "loss": 0.3819176483154297,
+      "step": 3800,
+      "token_acc": 0.9037970813391503
+    },
+    {
+      "epoch": 0.4167557170335542,
+      "grad_norm": 0.8828125,
+      "learning_rate": 3.2366151194720163e-06,
+      "loss": 0.4014278793334961,
+      "step": 3900,
+      "token_acc": 0.8989897804049096
+    },
+    {
+      "epoch": 0.42744176106005555,
+      "grad_norm": 1.2265625,
+      "learning_rate": 3.1543524772398464e-06,
+      "loss": 0.3903131103515625,
+      "step": 4000,
+      "token_acc": 0.9023779507523624
+    },
+    {
+      "epoch": 0.42744176106005555,
+      "eval_loss": 0.5559827089309692,
+      "eval_runtime": 1625.5851,
+      "eval_samples_per_second": 20.468,
+      "eval_steps_per_second": 1.28,
+      "eval_token_acc": 0.8759747254509199,
+      "step": 4000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 9358,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 2000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.218244349496107e+18,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

v0-20260123-014416/checkpoint-4000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e61a5ccecfe4850e11b9fb1b8bea0737456f069a42a114f94ef68317432e4e8
+size 7185

v0-20260123-014416/logging.jsonl

@@ -0,0 +1,44 @@
+{"loss": 0.71675009, "grad_norm": 1.6953125, "learning_rate": 3e-08, "token_acc": 0.84174877, "epoch": 0.00010686, "global_step/max_steps": "1/9358", "percentage": "0.01%", "elapsed_time": "11s", "remaining_time": "1d 6h 38m 27s", "memory(GiB)": 35.58, "train_speed(iter/s)": 0.084826}
+{"loss": 0.69730346, "grad_norm": 1.4453125, "learning_rate": 2.66e-06, "token_acc": 0.84671762, "epoch": 0.01068604, "global_step/max_steps": "100/9358", "percentage": "1.07%", "elapsed_time": "16m 25s", "remaining_time": "1d 1h 20m 48s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101459}
+{"loss": 0.6577813, "grad_norm": 1.046875, "learning_rate": 5e-06, "token_acc": 0.85318071, "epoch": 0.02137209, "global_step/max_steps": "200/9358", "percentage": "2.14%", "elapsed_time": "32m 50s", "remaining_time": "1d 1h 3m 50s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101495}
+{"loss": 0.60221527, "grad_norm": 1.328125, "learning_rate": 5e-06, "token_acc": 0.86209469, "epoch": 0.03205813, "global_step/max_steps": "300/9358", "percentage": "3.21%", "elapsed_time": "49m 14s", "remaining_time": "1d 0h 46m 52s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101533}
+{"loss": 0.55754822, "grad_norm": 1.2421875, "learning_rate": 4.99e-06, "token_acc": 0.86801399, "epoch": 0.04274418, "global_step/max_steps": "400/9358", "percentage": "4.27%", "elapsed_time": "1h 5m 36s", "remaining_time": "1d 0h 29m 27s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101603}
+{"loss": 0.52641785, "grad_norm": 1.8984375, "learning_rate": 4.99e-06, "token_acc": 0.86960294, "epoch": 0.05343022, "global_step/max_steps": "500/9358", "percentage": "5.34%", "elapsed_time": "1h 21m 59s", "remaining_time": "1d 0h 12m 32s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101638}
+{"loss": 0.47715324, "grad_norm": 1.1171875, "learning_rate": 4.98e-06, "token_acc": 0.87481657, "epoch": 0.06411626, "global_step/max_steps": "600/9358", "percentage": "6.41%", "elapsed_time": "1h 38m 27s", "remaining_time": "23h 57m 16s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101558}
+{"loss": 0.46431046, "grad_norm": 1.4765625, "learning_rate": 4.96e-06, "token_acc": 0.87540637, "epoch": 0.07480231, "global_step/max_steps": "700/9358", "percentage": "7.48%", "elapsed_time": "1h 54m 47s", "remaining_time": "23h 39m 42s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.10164}
+{"loss": 0.45615852, "grad_norm": 1.109375, "learning_rate": 4.95e-06, "token_acc": 0.8795579, "epoch": 0.08548835, "global_step/max_steps": "800/9358", "percentage": "8.55%", "elapsed_time": "2h 11m 18s", "remaining_time": "23h 24m 39s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101543}
+{"loss": 0.45342999, "grad_norm": 1.171875, "learning_rate": 4.93e-06, "token_acc": 0.88518798, "epoch": 0.0961744, "global_step/max_steps": "900/9358", "percentage": "9.62%", "elapsed_time": "2h 27m 38s", "remaining_time": "23h 7m 27s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101601}
+{"loss": 0.42084633, "grad_norm": 1.171875, "learning_rate": 4.9e-06, "token_acc": 0.89098993, "epoch": 0.10686044, "global_step/max_steps": "1000/9358", "percentage": "10.69%", "elapsed_time": "2h 44m 3s", "remaining_time": "22h 51m 9s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101593}
+{"loss": 0.43503849, "grad_norm": 1.390625, "learning_rate": 4.88e-06, "token_acc": 0.89003097, "epoch": 0.11754648, "global_step/max_steps": "1100/9358", "percentage": "11.75%", "elapsed_time": "3h 0m 28s", "remaining_time": "22h 34m 52s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101583}
+{"loss": 0.40780716, "grad_norm": 1.21875, "learning_rate": 4.85e-06, "token_acc": 0.89720866, "epoch": 0.12823253, "global_step/max_steps": "1200/9358", "percentage": "12.82%", "elapsed_time": "3h 16m 40s", "remaining_time": "22h 17m 4s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.10169}
+{"loss": 0.41691631, "grad_norm": 1.3046875, "learning_rate": 4.82e-06, "token_acc": 0.8944713, "epoch": 0.13891857, "global_step/max_steps": "1300/9358", "percentage": "13.89%", "elapsed_time": "3h 33m 2s", "remaining_time": "22h 0m 29s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101705}
+{"loss": 0.41801651, "grad_norm": 1.25, "learning_rate": 4.79e-06, "token_acc": 0.89545538, "epoch": 0.14960462, "global_step/max_steps": "1400/9358", "percentage": "14.96%", "elapsed_time": "3h 49m 21s", "remaining_time": "21h 43m 44s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101733}
+{"loss": 0.42521297, "grad_norm": 1.0390625, "learning_rate": 4.75e-06, "token_acc": 0.89274756, "epoch": 0.16029066, "global_step/max_steps": "1500/9358", "percentage": "16.03%", "elapsed_time": "4h 5m 54s", "remaining_time": "21h 28m 11s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101667}
+{"loss": 0.40242054, "grad_norm": 1.0625, "learning_rate": 4.71e-06, "token_acc": 0.89858694, "epoch": 0.1709767, "global_step/max_steps": "1600/9358", "percentage": "17.10%", "elapsed_time": "4h 22m 22s", "remaining_time": "21h 12m 12s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101635}
+{"loss": 0.39753117, "grad_norm": 1.6953125, "learning_rate": 4.67e-06, "token_acc": 0.90030837, "epoch": 0.18166275, "global_step/max_steps": "1700/9358", "percentage": "18.17%", "elapsed_time": "4h 38m 56s", "remaining_time": "20h 56m 35s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101572}
+{"loss": 0.4067078, "grad_norm": 1.328125, "learning_rate": 4.63e-06, "token_acc": 0.89851137, "epoch": 0.19234879, "global_step/max_steps": "1800/9358", "percentage": "19.23%", "elapsed_time": "4h 55m 21s", "remaining_time": "20h 40m 9s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101573}
+{"loss": 0.40403633, "grad_norm": 1.609375, "learning_rate": 4.58e-06, "token_acc": 0.89850942, "epoch": 0.20303484, "global_step/max_steps": "1900/9358", "percentage": "20.30%", "elapsed_time": "5h 11m 46s", "remaining_time": "20h 23m 47s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101569}
+{"loss": 0.38979122, "grad_norm": 1.3203125, "learning_rate": 4.53e-06, "token_acc": 0.90224994, "epoch": 0.21372088, "global_step/max_steps": "2000/9358", "percentage": "21.37%", "elapsed_time": "5h 28m 10s", "remaining_time": "20h 7m 20s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.101573}
+{"eval_loss": 0.55640507, "eval_runtime": 1628.334, "eval_samples_per_second": 20.433, "eval_steps_per_second": 1.277, "eval_token_acc": 0.87505222, "epoch": 0.21372088, "global_step/max_steps": "2000/9358", "percentage": "21.37%", "elapsed_time": "5h 55m 18s", "remaining_time": "21h 47m 11s", "memory(GiB)": 38.38, "train_speed(iter/s)": 0.093814}
+{"loss": 0.39870926, "grad_norm": 1.390625, "learning_rate": 4.48e-06, "token_acc": 0.899036, "epoch": 0.22440692, "global_step/max_steps": "2100/9358", "percentage": "22.44%", "elapsed_time": "6h 11m 49s", "remaining_time": "21h 25m 4s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.094132}
+{"loss": 0.38395836, "grad_norm": 1.140625, "learning_rate": 4.43e-06, "token_acc": 0.90277322, "epoch": 0.23509297, "global_step/max_steps": "2200/9358", "percentage": "23.51%", "elapsed_time": "6h 28m 24s", "remaining_time": "21h 3m 44s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.094402}
+{"loss": 0.39471344, "grad_norm": 0.96484375, "learning_rate": 4.37e-06, "token_acc": 0.90109563, "epoch": 0.24577901, "global_step/max_steps": "2300/9358", "percentage": "24.58%", "elapsed_time": "6h 45m 1s", "remaining_time": "20h 42m 52s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.094646}
+{"loss": 0.39796932, "grad_norm": 1.4140625, "learning_rate": 4.32e-06, "token_acc": 0.90076566, "epoch": 0.25646506, "global_step/max_steps": "2400/9358", "percentage": "25.65%", "elapsed_time": "7h 1m 23s", "remaining_time": "20h 21m 40s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.094924}
+{"loss": 0.3913583, "grad_norm": 1.265625, "learning_rate": 4.26e-06, "token_acc": 0.90150674, "epoch": 0.2671511, "global_step/max_steps": "2500/9358", "percentage": "26.72%", "elapsed_time": "7h 25m 44s", "remaining_time": "20h 22m 45s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.093477}
+{"loss": 0.38879887, "grad_norm": 1.46875, "learning_rate": 4.19e-06, "token_acc": 0.9019667, "epoch": 0.27783714, "global_step/max_steps": "2600/9358", "percentage": "27.78%", "elapsed_time": "7h 47m 49s", "remaining_time": "20h 15m 58s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.092628}
+{"loss": 0.39723473, "grad_norm": 2.4375, "learning_rate": 4.13e-06, "token_acc": 0.90014223, "epoch": 0.28852319, "global_step/max_steps": "2700/9358", "percentage": "28.85%", "elapsed_time": "8h 5m 56s", "remaining_time": "19h 58m 17s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.092604}
+{"loss": 0.3885001, "grad_norm": 1.2578125, "learning_rate": 4.06e-06, "token_acc": 0.90277885, "epoch": 0.29920923, "global_step/max_steps": "2800/9358", "percentage": "29.92%", "elapsed_time": "8h 28m 13s", "remaining_time": "19h 50m 19s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.091824}
+{"loss": 0.39520988, "grad_norm": 1.265625, "learning_rate": 4e-06, "token_acc": 0.90112791, "epoch": 0.30989528, "global_step/max_steps": "2900/9358", "percentage": "30.99%", "elapsed_time": "8h 53m 15s", "remaining_time": "19h 47m 30s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.090638}
+{"loss": 0.38712173, "grad_norm": 0.921875, "learning_rate": 3.93e-06, "token_acc": 0.90220833, "epoch": 0.32058132, "global_step/max_steps": "3000/9358", "percentage": "32.06%", "elapsed_time": "9h 19m 1s", "remaining_time": "19h 44m 44s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.089442}
+{"loss": 0.38103081, "grad_norm": 1.375, "learning_rate": 3.86e-06, "token_acc": 0.90346412, "epoch": 0.33126736, "global_step/max_steps": "3100/9358", "percentage": "33.13%", "elapsed_time": "9h 38m 26s", "remaining_time": "19h 27m 42s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.08932}
+{"loss": 0.37891117, "grad_norm": 1.5546875, "learning_rate": 3.78e-06, "token_acc": 0.90257153, "epoch": 0.34195341, "global_step/max_steps": "3200/9358", "percentage": "34.20%", "elapsed_time": "9h 54m 42s", "remaining_time": "19h 4m 27s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.089679}
+{"loss": 0.40724865, "grad_norm": 1.59375, "learning_rate": 3.71e-06, "token_acc": 0.89900145, "epoch": 0.35263945, "global_step/max_steps": "3300/9358", "percentage": "35.26%", "elapsed_time": "10h 15m 21s", "remaining_time": "18h 49m 38s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.089379}
+{"loss": 0.39273796, "grad_norm": 1.1171875, "learning_rate": 3.63e-06, "token_acc": 0.90093342, "epoch": 0.3633255, "global_step/max_steps": "3400/9358", "percentage": "36.33%", "elapsed_time": "10h 31m 45s", "remaining_time": "18h 27m 3s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.089697}
+{"loss": 0.39757305, "grad_norm": 1.5, "learning_rate": 3.56e-06, "token_acc": 0.89980192, "epoch": 0.37401154, "global_step/max_steps": "3500/9358", "percentage": "37.40%", "elapsed_time": "10h 48m 12s", "remaining_time": "18h 4m 54s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.089992}
+{"loss": 0.37646, "grad_norm": 2.140625, "learning_rate": 3.48e-06, "token_acc": 0.90510043, "epoch": 0.38469758, "global_step/max_steps": "3600/9358", "percentage": "38.47%", "elapsed_time": "11h 4m 32s", "remaining_time": "17h 42m 54s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.090287}
+{"loss": 0.41855862, "grad_norm": 1.1796875, "learning_rate": 3.4e-06, "token_acc": 0.8963375, "epoch": 0.39538363, "global_step/max_steps": "3700/9358", "percentage": "39.54%", "elapsed_time": "11h 33m 43s", "remaining_time": "17h 40m 50s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.088892}
+{"loss": 0.38191765, "grad_norm": 1.0390625, "learning_rate": 3.32e-06, "token_acc": 0.90379708, "epoch": 0.40606967, "global_step/max_steps": "3800/9358", "percentage": "40.61%", "elapsed_time": "12h 6m 24s", "remaining_time": "17h 42m 28s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.087187}
+{"loss": 0.40142788, "grad_norm": 0.8828125, "learning_rate": 3.24e-06, "token_acc": 0.89898978, "epoch": 0.41675572, "global_step/max_steps": "3900/9358", "percentage": "41.68%", "elapsed_time": "12h 44m 43s", "remaining_time": "17h 50m 13s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.084998}
+{"loss": 0.39031311, "grad_norm": 1.2265625, "learning_rate": 3.15e-06, "token_acc": 0.90237795, "epoch": 0.42744176, "global_step/max_steps": "4000/9358", "percentage": "42.74%", "elapsed_time": "13h 6m 19s", "remaining_time": "17h 33m 17s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.084782}
+{"eval_loss": 0.55598271, "eval_runtime": 1625.5851, "eval_samples_per_second": 20.468, "eval_steps_per_second": 1.28, "eval_token_acc": 0.87597473, "epoch": 0.42744176, "global_step/max_steps": "4000/9358", "percentage": "42.74%", "elapsed_time": "13h 33m 25s", "remaining_time": "18h 9m 35s", "memory(GiB)": 38.6, "train_speed(iter/s)": 0.081958}
+{"model_parameter_info": "PeftModelForCausalLM: 8321.1492M Params (5.9699M Trainable [0.0717%]), 0.0001M Buffers.", "last_model_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000", "best_model_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000", "best_metric": 0.55598271, "global_step": 4029, "log_history": [{"loss": 0.7167500853538513, "grad_norm": 1.6953125, "learning_rate": 2.6595744680851065e-08, "token_acc": 0.8417487684729064, "epoch": 0.00010686044026501389, "step": 1}, {"loss": 0.6973034637142913, "grad_norm": 1.4453125, "learning_rate": 2.6595744680851065e-06, "token_acc": 0.8467176205711577, "epoch": 0.010686044026501388, "step": 100}, {"loss": 0.6577812957763672, "grad_norm": 1.046875, "learning_rate": 4.999978873236658e-06, "token_acc": 0.8531807124137688, "epoch": 0.021372088053002777, "step": 200}, {"loss": 0.6022152709960937, "grad_norm": 1.328125, "learning_rate": 4.998159847367393e-06, "token_acc": 0.8620946875218537, "epoch": 0.03205813207950417, "step": 300}, {"loss": 0.5575482177734375, "grad_norm": 1.2421875, "learning_rate": 4.993408991262632e-06, "token_acc": 0.8680139909957781, "epoch": 0.04274417610600555, "step": 400}, {"loss": 0.5264178466796875, "grad_norm": 1.8984375, "learning_rate": 4.985731880507775e-06, "token_acc": 0.8696029378557931, "epoch": 0.05343022013250694, "step": 500}, {"loss": 0.4771532440185547, "grad_norm": 1.1171875, "learning_rate": 4.975137524928774e-06, "token_acc": 0.8748165713816137, "epoch": 0.06411626415900834, "step": 600}, {"loss": 0.4643104553222656, "grad_norm": 1.4765625, "learning_rate": 4.961638358018247e-06, "token_acc": 0.8754063671076119, "epoch": 0.07480230818550973, "step": 700}, {"loss": 0.4561585235595703, "grad_norm": 1.109375, "learning_rate": 4.945250222343571e-06, "token_acc": 0.8795579017210171, "epoch": 0.0854883522120111, "step": 800}, {"loss": 0.4534299850463867, "grad_norm": 1.171875, "learning_rate": 4.925992350954112e-06, "token_acc": 0.8851879828938838, "epoch": 0.0961743962385125, "step": 900}, {"loss": 0.42084632873535155, "grad_norm": 1.171875, "learning_rate": 4.903887344809403e-06, "token_acc": 0.8909899302321495, "epoch": 0.10686044026501389, "step": 1000}, {"loss": 0.43503849029541014, "grad_norm": 1.390625, "learning_rate": 4.8789611462547414e-06, "token_acc": 0.8900309663514769, "epoch": 0.11754648429151528, "step": 1100}, {"loss": 0.4078071594238281, "grad_norm": 1.21875, "learning_rate": 4.851243008575364e-06, "token_acc": 0.8972086577408744, "epoch": 0.12823252831801668, "step": 1200}, {"loss": 0.4169163131713867, "grad_norm": 1.3046875, "learning_rate": 4.8207654616649035e-06, "token_acc": 0.8944712980277317, "epoch": 0.13891857234451807, "step": 1300}, {"loss": 0.41801651000976564, "grad_norm": 1.25, "learning_rate": 4.7875642738484475e-06, "token_acc": 0.8954553844138678, "epoch": 0.14960461637101946, "step": 1400}, {"loss": 0.42521297454833984, "grad_norm": 1.0390625, "learning_rate": 4.7516784099049705e-06, "token_acc": 0.8927475600806661, "epoch": 0.16029066039752085, "step": 1500}, {"loss": 0.40242053985595705, "grad_norm": 1.0625, "learning_rate": 4.713149985338428e-06, "token_acc": 0.8985869393070196, "epoch": 0.1709767044240222, "step": 1600}, {"loss": 0.39753116607666017, "grad_norm": 1.6953125, "learning_rate": 4.672024216951176e-06, "token_acc": 0.9003083728795708, "epoch": 0.1816627484505236, "step": 1700}, {"loss": 0.40670780181884764, "grad_norm": 1.328125, "learning_rate": 4.628349369777712e-06, "token_acc": 0.8985113659223496, "epoch": 0.192348792477025, "step": 1800}, {"loss": 0.4040363311767578, "grad_norm": 1.609375, "learning_rate": 4.58217670044103e-06, "token_acc": 0.898509424415663, "epoch": 0.20303483650352638, "step": 1900}, {"loss": 0.3897912216186523, "grad_norm": 1.3203125, "learning_rate": 4.533560396998055e-06, "token_acc": 0.9022499382853919, "epoch": 0.21372088053002777, "step": 2000}, {"eval_loss": 0.5564050674438477, "eval_runtime": 1628.334, "eval_samples_per_second": 20.433, "eval_steps_per_second": 1.277, "eval_token_acc": 0.8750522224756502, "epoch": 0.21372088053002777, "step": 2000}, {"loss": 0.39870925903320314, "grad_norm": 1.390625, "learning_rate": 4.482557515344757e-06, "token_acc": 0.8990360044410765, "epoch": 0.22440692455652916, "step": 2100}, {"loss": 0.38395835876464846, "grad_norm": 1.140625, "learning_rate": 4.429227912255589e-06, "token_acc": 0.902773223403757, "epoch": 0.23509296858303055, "step": 2200}, {"loss": 0.39471343994140623, "grad_norm": 0.96484375, "learning_rate": 4.3736341751358166e-06, "token_acc": 0.9010956332106893, "epoch": 0.24577901260953194, "step": 2300}, {"loss": 0.39796932220458986, "grad_norm": 1.4140625, "learning_rate": 4.315841548569193e-06, "token_acc": 0.9007656570069128, "epoch": 0.25646505663603336, "step": 2400}, {"loss": 0.3913582992553711, "grad_norm": 1.265625, "learning_rate": 4.255917857747183e-06, "token_acc": 0.9015067434468572, "epoch": 0.2671511006625347, "step": 2500}, {"loss": 0.38879886627197263, "grad_norm": 1.46875, "learning_rate": 4.193933428869592e-06, "token_acc": 0.9019667027692453, "epoch": 0.27783714468903614, "step": 2600}, {"loss": 0.3972347259521484, "grad_norm": 2.4375, "learning_rate": 4.129961006610028e-06, "token_acc": 0.9001422292689183, "epoch": 0.2885231887155375, "step": 2700}, {"loss": 0.3885000991821289, "grad_norm": 1.2578125, "learning_rate": 4.064075668743043e-06, "token_acc": 0.9027788505528588, "epoch": 0.2992092327420389, "step": 2800}, {"loss": 0.39520988464355467, "grad_norm": 1.265625, "learning_rate": 3.996354738033166e-06, "token_acc": 0.9011279097672186, "epoch": 0.3098952767685403, "step": 2900}, {"loss": 0.3871217346191406, "grad_norm": 0.921875, "learning_rate": 3.926877691489222e-06, "token_acc": 0.9022083313787118, "epoch": 0.3205813207950417, "step": 3000}, {"loss": 0.3810308074951172, "grad_norm": 1.375, "learning_rate": 3.855726067090433e-06, "token_acc": 0.9034641205430405, "epoch": 0.33126736482154306, "step": 3100}, {"loss": 0.37891117095947263, "grad_norm": 1.5546875, "learning_rate": 3.7829833680937804e-06, "token_acc": 0.9025715320536037, "epoch": 0.3419534088480444, "step": 3200}, {"loss": 0.40724864959716794, "grad_norm": 1.59375, "learning_rate": 3.7087349650349234e-06, "token_acc": 0.8990014549892049, "epoch": 0.35263945287454584, "step": 3300}, {"loss": 0.39273796081542967, "grad_norm": 1.1171875, "learning_rate": 3.6330679955376843e-06, "token_acc": 0.900933421493117, "epoch": 0.3633254969010472, "step": 3400}, {"loss": 0.3975730514526367, "grad_norm": 1.5, "learning_rate": 3.556071262049684e-06, "token_acc": 0.8998019157822584, "epoch": 0.3740115409275486, "step": 3500}, {"loss": 0.37645999908447264, "grad_norm": 2.140625, "learning_rate": 3.477835127624158e-06, "token_acc": 0.9051004274755519, "epoch": 0.38469758495405, "step": 3600}, {"loss": 0.4185586166381836, "grad_norm": 1.1796875, "learning_rate": 3.39845140987024e-06, "token_acc": 0.8963374986700084, "epoch": 0.3953836289805514, "step": 3700}, {"loss": 0.3819176483154297, "grad_norm": 1.0390625, "learning_rate": 3.318013273196187e-06, "token_acc": 0.9037970813391503, "epoch": 0.40606967300705277, "step": 3800}, {"loss": 0.4014278793334961, "grad_norm": 0.8828125, "learning_rate": 3.2366151194720163e-06, "token_acc": 0.8989897804049096, "epoch": 0.4167557170335542, "step": 3900}, {"loss": 0.3903131103515625, "grad_norm": 1.2265625, "learning_rate": 3.1543524772398464e-06, "token_acc": 0.9023779507523624, "epoch": 0.42744176106005555, "step": 4000}, {"eval_loss": 0.5559827089309692, "eval_runtime": 1625.5851, "eval_samples_per_second": 20.468, "eval_steps_per_second": 1.28, "eval_token_acc": 0.8759747254509199, "epoch": 0.42744176106005555, "step": 4000}], "memory": 38.603515625}

v0-20260123-014416/v0-20260123-194351/args.json

@@ -0,0 +1,351 @@
+{
+  "output_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v0-20260123-194351",
+  "overwrite_output_dir": false,
+  "do_train": false,
+  "do_eval": false,
+  "do_predict": false,
+  "eval_strategy": "steps",
+  "prediction_loss_only": false,
+  "per_device_train_batch_size": 16,
+  "per_device_eval_batch_size": 16,
+  "per_gpu_train_batch_size": null,
+  "per_gpu_eval_batch_size": null,
+  "gradient_accumulation_steps": 4,
+  "eval_accumulation_steps": null,
+  "eval_delay": 0,
+  "torch_empty_cache_steps": null,
+  "learning_rate": 5e-06,
+  "weight_decay": 0.01,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "max_grad_norm": 1.0,
+  "num_train_epochs": 1.0,
+  "max_steps": -1,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": null,
+  "warmup_ratio": 0.02,
+  "warmup_steps": 0,
+  "log_level": "passive",
+  "log_level_replica": "warning",
+  "log_on_each_node": true,
+  "logging_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v0-20260123-194351/runs",
+  "logging_strategy": "steps",
+  "logging_first_step": true,
+  "logging_steps": 100,
+  "logging_nan_inf_filter": true,
+  "save_strategy": "steps",
+  "save_steps": 2000.0,
+  "save_total_limit": 40,
+  "save_safetensors": true,
+  "save_on_each_node": false,
+  "save_only_model": false,
+  "restore_callback_states_from_checkpoint": false,
+  "no_cuda": false,
+  "use_cpu": false,
+  "use_mps_device": false,
+  "seed": 42,
+  "data_seed": 42,
+  "jit_mode_eval": false,
+  "use_ipex": false,
+  "bf16": true,
+  "fp16": false,
+  "fp16_opt_level": "O1",
+  "half_precision_backend": "auto",
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "local_rank": -1,
+  "ddp_backend": null,
+  "tpu_num_cores": null,
+  "tpu_metrics_debug": false,
+  "debug": null,
+  "dataloader_drop_last": false,
+  "eval_steps": 2000.0,
+  "dataloader_num_workers": null,
+  "dataloader_prefetch_factor": null,
+  "past_index": -1,
+  "run_name": "step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF",
+  "disable_tqdm": null,
+  "remove_unused_columns": true,
+  "label_names": null,
+  "load_best_model_at_end": false,
+  "metric_for_best_model": "loss",
+  "greater_is_better": false,
+  "ignore_data_skip": false,
+  "fsdp": "",
+  "fsdp_min_num_params": 0,
+  "fsdp_config": null,
+  "fsdp_transformer_layer_cls_to_wrap": null,
+  "accelerator_config": {
+    "dispatch_batches": false
+  },
+  "deepspeed": null,
+  "label_smoothing_factor": 0.0,
+  "optim": "adamw_torch",
+  "optim_args": null,
+  "adafactor": false,
+  "group_by_length": false,
+  "length_column_name": "length",
+  "report_to": [
+    "wandb"
+  ],
+  "ddp_find_unused_parameters": null,
+  "ddp_bucket_cap_mb": null,
+  "ddp_broadcast_buffers": null,
+  "dataloader_pin_memory": true,
+  "dataloader_persistent_workers": false,
+  "skip_memory_metrics": true,
+  "use_legacy_prediction_loop": false,
+  "push_to_hub": false,
+  "resume_from_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000",
+  "hub_model_id": null,
+  "hub_strategy": "every_save",
+  "hub_token": null,
+  "hub_private_repo": null,
+  "hub_always_push": false,
+  "gradient_checkpointing": true,
+  "gradient_checkpointing_kwargs": "{\"use_reentrant\": false}",
+  "include_inputs_for_metrics": false,
+  "include_for_metrics": [],
+  "eval_do_concat_batches": true,
+  "fp16_backend": "auto",
+  "evaluation_strategy": "steps",
+  "push_to_hub_model_id": null,
+  "push_to_hub_organization": null,
+  "push_to_hub_token": null,
+  "mp_parameters": "",
+  "auto_find_batch_size": false,
+  "full_determinism": false,
+  "torchdynamo": null,
+  "ray_scope": "last",
+  "ddp_timeout": 18000000,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null,
+  "dispatch_batches": null,
+  "split_batches": null,
+  "include_tokens_per_second": false,
+  "include_num_input_tokens_seen": false,
+  "neftune_noise_alpha": null,
+  "optim_target_modules": null,
+  "batch_eval_metrics": false,
+  "eval_on_start": false,
+  "use_liger_kernel": false,
+  "eval_use_gather_object": false,
+  "average_tokens_across_devices": false,
+  "sortish_sampler": false,
+  "predict_with_generate": false,
+  "generation_max_length": null,
+  "generation_num_beams": null,
+  "generation_config": null,
+  "tuner_backend": "peft",
+  "vit_gradient_checkpointing": null,
+  "router_aux_loss_coef": 0.0,
+  "enable_dft_loss": false,
+  "enable_channel_loss": false,
+  "check_model": true,
+  "acc_strategy": "token",
+  "train_dataloader_shuffle": true,
+  "max_epochs": null,
+  "aligner_lr": 1e-05,
+  "vit_lr": 5e-06,
+  "use_logits_to_keep": null,
+  "ds3_gather_for_generation": true,
+  "resume_only_model": false,
+  "optimizer": null,
+  "loss_type": null,
+  "metric": null,
+  "eval_use_evalscope": false,
+  "eval_dataset": [],
+  "eval_dataset_args": null,
+  "eval_limit": null,
+  "eval_generation_config": null,
+  "extra_eval_args": null,
+  "use_flash_ckpt": false,
+  "use_ray": false,
+  "ray_exp_name": null,
+  "device_groups": null,
+  "model": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "model_type": "step_audio2_mini",
+  "model_revision": null,
+  "task_type": "causal_lm",
+  "torch_dtype": "bfloat16",
+  "attn_impl": null,
+  "new_special_tokens": [],
+  "num_labels": null,
+  "problem_type": null,
+  "rope_scaling": null,
+  "device_map": null,
+  "max_memory": {},
+  "max_model_len": null,
+  "local_repo_path": null,
+  "init_strategy": null,
+  "template": "step_audio2_mini",
+  "system": null,
+  "max_length": 16384,
+  "truncation_strategy": "delete",
+  "max_pixels": null,
+  "agent_template": null,
+  "norm_bbox": null,
+  "use_chat_template": true,
+  "padding_free": false,
+  "padding_side": "right",
+  "loss_scale": "default",
+  "sequence_parallel_size": 1,
+  "response_prefix": null,
+  "template_backend": "swift",
+  "dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_train90.jsonl"
+  ],
+  "val_dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "cached_dataset": [],
+  "cached_val_dataset": [],
+  "split_dataset_ratio": 0.0,
+  "dataset_num_proc": 1,
+  "load_from_cache_file": false,
+  "dataset_shuffle": true,
+  "val_dataset_shuffle": false,
+  "streaming": false,
+  "interleave_prob": null,
+  "stopping_strategy": "first_exhausted",
+  "shuffle_buffer_size": 1000,
+  "download_mode": "reuse_dataset_if_exists",
+  "columns": {},
+  "strict": false,
+  "model_name": null,
+  "model_author": null,
+  "custom_dataset_info": [],
+  "quant_method": null,
+  "quant_bits": null,
+  "hqq_axis": null,
+  "bnb_4bit_compute_dtype": "bfloat16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "max_new_tokens": 64,
+  "temperature": 0.0,
+  "top_k": null,
+  "top_p": null,
+  "repetition_penalty": null,
+  "num_beams": 1,
+  "stream": false,
+  "stop_words": [],
+  "logprobs": false,
+  "top_logprobs": null,
+  "ckpt_dir": null,
+  "lora_modules": [],
+  "train_type": "lora",
+  "adapters": [],
+  "external_plugins": [],
+  "model_kwargs": {},
+  "load_args": false,
+  "load_data_args": false,
+  "packing": false,
+  "packing_length": null,
+  "packing_num_proc": 1,
+  "lazy_tokenize": true,
+  "custom_register_path": [],
+  "use_hf": false,
+  "ignore_args_error": false,
+  "use_swift_lora": false,
+  "freeze_parameters": [],
+  "freeze_parameters_regex": null,
+  "freeze_parameters_ratio": 0.0,
+  "trainable_parameters": [],
+  "trainable_parameters_regex": null,
+  "freeze_llm": true,
+  "freeze_vit": false,
+  "freeze_aligner": false,
+  "target_modules": [
+    "all-linear"
+  ],
+  "target_regex": null,
+  "target_parameters": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "lora_bias": "none",
+  "lora_dtype": null,
+  "lorap_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "lora_ga_batch_size": 2,
+  "lora_ga_iters": 2,
+  "lora_ga_max_length": 1024,
+  "lora_ga_direction": "ArB2r",
+  "lora_ga_scale": "stable",
+  "lora_ga_stable_gamma": 16,
+  "init_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "use_galore": false,
+  "galore_target_modules": null,
+  "galore_rank": 128,
+  "galore_update_proj_gap": 50,
+  "galore_scale": 1.0,
+  "galore_proj_type": "std",
+  "galore_optim_per_parameter": false,
+  "galore_with_embedding": false,
+  "galore_quantization": false,
+  "galore_proj_quant": false,
+  "galore_proj_bits": 4,
+  "galore_proj_group_size": 256,
+  "galore_cos_threshold": 0.4,
+  "galore_gamma_proj": 2,
+  "galore_queue_size": 5,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "lisa_activated_layers": 0,
+  "lisa_step_interval": 20,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "swanlab_token": null,
+  "swanlab_project": null,
+  "swanlab_workspace": null,
+  "swanlab_exp_name": null,
+  "swanlab_lark_webhook_url": null,
+  "swanlab_lark_secret": null,
+  "swanlab_mode": "cloud",
+  "add_version": true,
+  "create_checkpoint_symlink": false,
+  "zero_hpz_partition_size": null,
+  "deepspeed_autotp_size": null,
+  "early_stop_interval": null,
+  "rank": -1,
+  "global_world_size": 1,
+  "local_world_size": 1,
+  "model_suffix": "Step-Audio-2-mini",
+  "model_info": "ModelInfo(model_type='step_audio2_mini', model_dir='/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini', torch_dtype=torch.bfloat16, max_model_len=16384, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=True, config=None, task_type='causal_lm', num_labels=None)",
+  "model_meta": "ModelMeta(model_type='step_audio2_mini', model_groups=[ModelGroup(models=[Model(ms_model_id='stepfun-ai/Step-Audio-2-mini', hf_model_id='stepfun-ai/Step-Audio-2-mini', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='step_audio2_mini', get_function=<function get_model_tokenizer_step_audio2_mini at 0x7f3879f384c0>, model_arch=MultiModelKeys(arch_name='step_audio2_mini', embedding=None, module_list=None, lm_head=None, q_proj=None, k_proj=None, v_proj=None, o_proj=None, attention=None, mlp=None, down_proj=None, qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None, language_model=['model', 'lm_head'], aligner=['adapter'], vision_tower=['encoder'], generator=[]), architectures=['StepAudio2ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=True, is_reward=False, is_reranker=False, task_type=None, ignore_patterns=None, requires=['transformers==4.53.3', 'torchaudio', 'librosa'], tags=['audio'])",
+  "model_dir": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "_val_dataset_exists": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "hub": "<class 'swift.hub.hub.MSHub'>",
+  "training_args": "Seq2SeqTrainingArguments(output_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v0-20260123-194351', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=16, per_device_eval_batch_size=16, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=4, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=5e-06, weight_decay=0.01, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=1.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.02, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v0-20260123-194351/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=100, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=2000, save_total_limit=40, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=2000, dataloader_num_workers=1, dataloader_prefetch_factor=10, past_index=-1, run_name='step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['wandb'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000', hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs={'use_reentrant': False}, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy='steps', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, eval_use_gather_object=False, average_tokens_across_devices=None, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=1e-05, vit_lr=5e-06, use_logits_to_keep=None, ds3_gather_for_generation=True, resume_only_model=False, optimizer='multimodal', loss_type=None, metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='lora', local_repo_path=None, galore_config=None, padding_side='right', padding_free=False, task_type='causal_lm', problem_type=None)"
+}

v0-20260123-014416/v0-20260123-194351/logging.jsonl

@@ -0,0 +1 @@
+{"model_parameter_info": "PeftModelForCausalLM: 8321.1492M Params (5.9699M Trainable [0.0717%]), 0.0001M Buffers.", "last_model_checkpoint": null, "best_model_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000", "best_metric": 0.55598271, "global_step": 4029, "log_history": [{"epoch": 0.00010686044026501389, "grad_norm": 1.6953125, "learning_rate": 2.6595744680851065e-08, "loss": 0.7167500853538513, "step": 1, "token_acc": 0.8417487684729064}, {"epoch": 0.010686044026501388, "grad_norm": 1.4453125, "learning_rate": 2.6595744680851065e-06, "loss": 0.6973034637142913, "step": 100, "token_acc": 0.8467176205711577}, {"epoch": 0.021372088053002777, "grad_norm": 1.046875, "learning_rate": 4.999978873236658e-06, "loss": 0.6577812957763672, "step": 200, "token_acc": 0.8531807124137688}, {"epoch": 0.03205813207950417, "grad_norm": 1.328125, "learning_rate": 4.998159847367393e-06, "loss": 0.6022152709960937, "step": 300, "token_acc": 0.8620946875218537}, {"epoch": 0.04274417610600555, "grad_norm": 1.2421875, "learning_rate": 4.993408991262632e-06, "loss": 0.5575482177734375, "step": 400, "token_acc": 0.8680139909957781}, {"epoch": 0.05343022013250694, "grad_norm": 1.8984375, "learning_rate": 4.985731880507775e-06, "loss": 0.5264178466796875, "step": 500, "token_acc": 0.8696029378557931}, {"epoch": 0.06411626415900834, "grad_norm": 1.1171875, "learning_rate": 4.975137524928774e-06, "loss": 0.4771532440185547, "step": 600, "token_acc": 0.8748165713816137}, {"epoch": 0.07480230818550973, "grad_norm": 1.4765625, "learning_rate": 4.961638358018247e-06, "loss": 0.4643104553222656, "step": 700, "token_acc": 0.8754063671076119}, {"epoch": 0.0854883522120111, "grad_norm": 1.109375, "learning_rate": 4.945250222343571e-06, "loss": 0.4561585235595703, "step": 800, "token_acc": 0.8795579017210171}, {"epoch": 0.0961743962385125, "grad_norm": 1.171875, "learning_rate": 4.925992350954112e-06, "loss": 0.4534299850463867, "step": 900, "token_acc": 0.8851879828938838}, {"epoch": 0.10686044026501389, "grad_norm": 1.171875, "learning_rate": 4.903887344809403e-06, "loss": 0.42084632873535155, "step": 1000, "token_acc": 0.8909899302321495}, {"epoch": 0.11754648429151528, "grad_norm": 1.390625, "learning_rate": 4.8789611462547414e-06, "loss": 0.43503849029541014, "step": 1100, "token_acc": 0.8900309663514769}, {"epoch": 0.12823252831801668, "grad_norm": 1.21875, "learning_rate": 4.851243008575364e-06, "loss": 0.4078071594238281, "step": 1200, "token_acc": 0.8972086577408744}, {"epoch": 0.13891857234451807, "grad_norm": 1.3046875, "learning_rate": 4.8207654616649035e-06, "loss": 0.4169163131713867, "step": 1300, "token_acc": 0.8944712980277317}, {"epoch": 0.14960461637101946, "grad_norm": 1.25, "learning_rate": 4.7875642738484475e-06, "loss": 0.41801651000976564, "step": 1400, "token_acc": 0.8954553844138678}, {"epoch": 0.16029066039752085, "grad_norm": 1.0390625, "learning_rate": 4.7516784099049705e-06, "loss": 0.42521297454833984, "step": 1500, "token_acc": 0.8927475600806661}, {"epoch": 0.1709767044240222, "grad_norm": 1.0625, "learning_rate": 4.713149985338428e-06, "loss": 0.40242053985595705, "step": 1600, "token_acc": 0.8985869393070196}, {"epoch": 0.1816627484505236, "grad_norm": 1.6953125, "learning_rate": 4.672024216951176e-06, "loss": 0.39753116607666017, "step": 1700, "token_acc": 0.9003083728795708}, {"epoch": 0.192348792477025, "grad_norm": 1.328125, "learning_rate": 4.628349369777712e-06, "loss": 0.40670780181884764, "step": 1800, "token_acc": 0.8985113659223496}, {"epoch": 0.20303483650352638, "grad_norm": 1.609375, "learning_rate": 4.58217670044103e-06, "loss": 0.4040363311767578, "step": 1900, "token_acc": 0.898509424415663}, {"epoch": 0.21372088053002777, "grad_norm": 1.3203125, "learning_rate": 4.533560396998055e-06, "loss": 0.3897912216186523, "step": 2000, "token_acc": 0.9022499382853919}, {"epoch": 0.21372088053002777, "eval_loss": 0.5564050674438477, "eval_runtime": 1628.334, "eval_samples_per_second": 20.433, "eval_steps_per_second": 1.277, "eval_token_acc": 0.8750522224756502, "step": 2000}, {"epoch": 0.22440692455652916, "grad_norm": 1.390625, "learning_rate": 4.482557515344757e-06, "loss": 0.39870925903320314, "step": 2100, "token_acc": 0.8990360044410765}, {"epoch": 0.23509296858303055, "grad_norm": 1.140625, "learning_rate": 4.429227912255589e-06, "loss": 0.38395835876464846, "step": 2200, "token_acc": 0.902773223403757}, {"epoch": 0.24577901260953194, "grad_norm": 0.96484375, "learning_rate": 4.3736341751358166e-06, "loss": 0.39471343994140623, "step": 2300, "token_acc": 0.9010956332106893}, {"epoch": 0.25646505663603336, "grad_norm": 1.4140625, "learning_rate": 4.315841548569193e-06, "loss": 0.39796932220458986, "step": 2400, "token_acc": 0.9007656570069128}, {"epoch": 0.2671511006625347, "grad_norm": 1.265625, "learning_rate": 4.255917857747183e-06, "loss": 0.3913582992553711, "step": 2500, "token_acc": 0.9015067434468572}, {"epoch": 0.27783714468903614, "grad_norm": 1.46875, "learning_rate": 4.193933428869592e-06, "loss": 0.38879886627197263, "step": 2600, "token_acc": 0.9019667027692453}, {"epoch": 0.2885231887155375, "grad_norm": 2.4375, "learning_rate": 4.129961006610028e-06, "loss": 0.3972347259521484, "step": 2700, "token_acc": 0.9001422292689183}, {"epoch": 0.2992092327420389, "grad_norm": 1.2578125, "learning_rate": 4.064075668743043e-06, "loss": 0.3885000991821289, "step": 2800, "token_acc": 0.9027788505528588}, {"epoch": 0.3098952767685403, "grad_norm": 1.265625, "learning_rate": 3.996354738033166e-06, "loss": 0.39520988464355467, "step": 2900, "token_acc": 0.9011279097672186}, {"epoch": 0.3205813207950417, "grad_norm": 0.921875, "learning_rate": 3.926877691489222e-06, "loss": 0.3871217346191406, "step": 3000, "token_acc": 0.9022083313787118}, {"epoch": 0.33126736482154306, "grad_norm": 1.375, "learning_rate": 3.855726067090433e-06, "loss": 0.3810308074951172, "step": 3100, "token_acc": 0.9034641205430405}, {"epoch": 0.3419534088480444, "grad_norm": 1.5546875, "learning_rate": 3.7829833680937804e-06, "loss": 0.37891117095947263, "step": 3200, "token_acc": 0.9025715320536037}, {"epoch": 0.35263945287454584, "grad_norm": 1.59375, "learning_rate": 3.7087349650349234e-06, "loss": 0.40724864959716794, "step": 3300, "token_acc": 0.8990014549892049}, {"epoch": 0.3633254969010472, "grad_norm": 1.1171875, "learning_rate": 3.6330679955376843e-06, "loss": 0.39273796081542967, "step": 3400, "token_acc": 0.900933421493117}, {"epoch": 0.3740115409275486, "grad_norm": 1.5, "learning_rate": 3.556071262049684e-06, "loss": 0.3975730514526367, "step": 3500, "token_acc": 0.8998019157822584}, {"epoch": 0.38469758495405, "grad_norm": 2.140625, "learning_rate": 3.477835127624158e-06, "loss": 0.37645999908447264, "step": 3600, "token_acc": 0.9051004274755519}, {"epoch": 0.3953836289805514, "grad_norm": 1.1796875, "learning_rate": 3.39845140987024e-06, "loss": 0.4185586166381836, "step": 3700, "token_acc": 0.8963374986700084}, {"epoch": 0.40606967300705277, "grad_norm": 1.0390625, "learning_rate": 3.318013273196187e-06, "loss": 0.3819176483154297, "step": 3800, "token_acc": 0.9037970813391503}, {"epoch": 0.4167557170335542, "grad_norm": 0.8828125, "learning_rate": 3.2366151194720163e-06, "loss": 0.4014278793334961, "step": 3900, "token_acc": 0.8989897804049096}, {"epoch": 0.42744176106005555, "grad_norm": 1.2265625, "learning_rate": 3.1543524772398464e-06, "loss": 0.3903131103515625, "step": 4000, "token_acc": 0.9023779507523624}, {"epoch": 0.42744176106005555, "eval_loss": 0.5559827089309692, "eval_runtime": 1625.5851, "eval_samples_per_second": 20.468, "eval_steps_per_second": 1.28, "eval_token_acc": 0.8759747254509199, "step": 4000}], "memory": null}

v0-20260123-014416/v1-20260123-200453/args.json

@@ -0,0 +1,351 @@
+{
+  "output_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453",
+  "overwrite_output_dir": false,
+  "do_train": false,
+  "do_eval": false,
+  "do_predict": false,
+  "eval_strategy": "steps",
+  "prediction_loss_only": false,
+  "per_device_train_batch_size": 16,
+  "per_device_eval_batch_size": 16,
+  "per_gpu_train_batch_size": null,
+  "per_gpu_eval_batch_size": null,
+  "gradient_accumulation_steps": 4,
+  "eval_accumulation_steps": null,
+  "eval_delay": 0,
+  "torch_empty_cache_steps": null,
+  "learning_rate": 5e-06,
+  "weight_decay": 0.01,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "max_grad_norm": 1.0,
+  "num_train_epochs": 1.0,
+  "max_steps": -1,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": null,
+  "warmup_ratio": 0.02,
+  "warmup_steps": 0,
+  "log_level": "passive",
+  "log_level_replica": "warning",
+  "log_on_each_node": true,
+  "logging_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453/runs",
+  "logging_strategy": "steps",
+  "logging_first_step": true,
+  "logging_steps": 100,
+  "logging_nan_inf_filter": true,
+  "save_strategy": "steps",
+  "save_steps": 2000.0,
+  "save_total_limit": 40,
+  "save_safetensors": true,
+  "save_on_each_node": false,
+  "save_only_model": false,
+  "restore_callback_states_from_checkpoint": false,
+  "no_cuda": false,
+  "use_cpu": false,
+  "use_mps_device": false,
+  "seed": 42,
+  "data_seed": 42,
+  "jit_mode_eval": false,
+  "use_ipex": false,
+  "bf16": true,
+  "fp16": false,
+  "fp16_opt_level": "O1",
+  "half_precision_backend": "auto",
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "local_rank": -1,
+  "ddp_backend": null,
+  "tpu_num_cores": null,
+  "tpu_metrics_debug": false,
+  "debug": null,
+  "dataloader_drop_last": false,
+  "eval_steps": 2000.0,
+  "dataloader_num_workers": null,
+  "dataloader_prefetch_factor": null,
+  "past_index": -1,
+  "run_name": "step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF",
+  "disable_tqdm": null,
+  "remove_unused_columns": true,
+  "label_names": null,
+  "load_best_model_at_end": false,
+  "metric_for_best_model": "loss",
+  "greater_is_better": false,
+  "ignore_data_skip": false,
+  "fsdp": "",
+  "fsdp_min_num_params": 0,
+  "fsdp_config": null,
+  "fsdp_transformer_layer_cls_to_wrap": null,
+  "accelerator_config": {
+    "dispatch_batches": false
+  },
+  "deepspeed": null,
+  "label_smoothing_factor": 0.0,
+  "optim": "adamw_torch",
+  "optim_args": null,
+  "adafactor": false,
+  "group_by_length": false,
+  "length_column_name": "length",
+  "report_to": [
+    "wandb"
+  ],
+  "ddp_find_unused_parameters": null,
+  "ddp_bucket_cap_mb": null,
+  "ddp_broadcast_buffers": null,
+  "dataloader_pin_memory": true,
+  "dataloader_persistent_workers": false,
+  "skip_memory_metrics": true,
+  "use_legacy_prediction_loop": false,
+  "push_to_hub": false,
+  "resume_from_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000",
+  "hub_model_id": null,
+  "hub_strategy": "every_save",
+  "hub_token": null,
+  "hub_private_repo": null,
+  "hub_always_push": false,
+  "gradient_checkpointing": true,
+  "gradient_checkpointing_kwargs": "{\"use_reentrant\": false}",
+  "include_inputs_for_metrics": false,
+  "include_for_metrics": [],
+  "eval_do_concat_batches": true,
+  "fp16_backend": "auto",
+  "evaluation_strategy": "steps",
+  "push_to_hub_model_id": null,
+  "push_to_hub_organization": null,
+  "push_to_hub_token": null,
+  "mp_parameters": "",
+  "auto_find_batch_size": false,
+  "full_determinism": false,
+  "torchdynamo": null,
+  "ray_scope": "last",
+  "ddp_timeout": 18000000,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null,
+  "dispatch_batches": null,
+  "split_batches": null,
+  "include_tokens_per_second": false,
+  "include_num_input_tokens_seen": false,
+  "neftune_noise_alpha": null,
+  "optim_target_modules": null,
+  "batch_eval_metrics": false,
+  "eval_on_start": false,
+  "use_liger_kernel": false,
+  "eval_use_gather_object": false,
+  "average_tokens_across_devices": false,
+  "sortish_sampler": false,
+  "predict_with_generate": false,
+  "generation_max_length": null,
+  "generation_num_beams": null,
+  "generation_config": null,
+  "tuner_backend": "peft",
+  "vit_gradient_checkpointing": null,
+  "router_aux_loss_coef": 0.0,
+  "enable_dft_loss": false,
+  "enable_channel_loss": false,
+  "check_model": true,
+  "acc_strategy": "token",
+  "train_dataloader_shuffle": true,
+  "max_epochs": null,
+  "aligner_lr": 1e-05,
+  "vit_lr": 5e-06,
+  "use_logits_to_keep": null,
+  "ds3_gather_for_generation": true,
+  "resume_only_model": false,
+  "optimizer": null,
+  "loss_type": null,
+  "metric": null,
+  "eval_use_evalscope": false,
+  "eval_dataset": [],
+  "eval_dataset_args": null,
+  "eval_limit": null,
+  "eval_generation_config": null,
+  "extra_eval_args": null,
+  "use_flash_ckpt": false,
+  "use_ray": false,
+  "ray_exp_name": null,
+  "device_groups": null,
+  "model": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "model_type": "step_audio2_mini",
+  "model_revision": null,
+  "task_type": "causal_lm",
+  "torch_dtype": "bfloat16",
+  "attn_impl": null,
+  "new_special_tokens": [],
+  "num_labels": null,
+  "problem_type": null,
+  "rope_scaling": null,
+  "device_map": null,
+  "max_memory": {},
+  "max_model_len": null,
+  "local_repo_path": null,
+  "init_strategy": null,
+  "template": "step_audio2_mini",
+  "system": null,
+  "max_length": 16384,
+  "truncation_strategy": "delete",
+  "max_pixels": null,
+  "agent_template": null,
+  "norm_bbox": null,
+  "use_chat_template": true,
+  "padding_free": false,
+  "padding_side": "right",
+  "loss_scale": "default",
+  "sequence_parallel_size": 1,
+  "response_prefix": null,
+  "template_backend": "swift",
+  "dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_train90.jsonl"
+  ],
+  "val_dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "cached_dataset": [],
+  "cached_val_dataset": [],
+  "split_dataset_ratio": 0.0,
+  "dataset_num_proc": 1,
+  "load_from_cache_file": false,
+  "dataset_shuffle": true,
+  "val_dataset_shuffle": false,
+  "streaming": false,
+  "interleave_prob": null,
+  "stopping_strategy": "first_exhausted",
+  "shuffle_buffer_size": 1000,
+  "download_mode": "reuse_dataset_if_exists",
+  "columns": {},
+  "strict": false,
+  "model_name": null,
+  "model_author": null,
+  "custom_dataset_info": [],
+  "quant_method": null,
+  "quant_bits": null,
+  "hqq_axis": null,
+  "bnb_4bit_compute_dtype": "bfloat16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "max_new_tokens": 64,
+  "temperature": 0.0,
+  "top_k": null,
+  "top_p": null,
+  "repetition_penalty": null,
+  "num_beams": 1,
+  "stream": false,
+  "stop_words": [],
+  "logprobs": false,
+  "top_logprobs": null,
+  "ckpt_dir": null,
+  "lora_modules": [],
+  "train_type": "lora",
+  "adapters": [],
+  "external_plugins": [],
+  "model_kwargs": {},
+  "load_args": false,
+  "load_data_args": false,
+  "packing": false,
+  "packing_length": null,
+  "packing_num_proc": 1,
+  "lazy_tokenize": true,
+  "custom_register_path": [],
+  "use_hf": false,
+  "ignore_args_error": false,
+  "use_swift_lora": false,
+  "freeze_parameters": [],
+  "freeze_parameters_regex": null,
+  "freeze_parameters_ratio": 0.0,
+  "trainable_parameters": [],
+  "trainable_parameters_regex": null,
+  "freeze_llm": true,
+  "freeze_vit": false,
+  "freeze_aligner": false,
+  "target_modules": [
+    "all-linear"
+  ],
+  "target_regex": null,
+  "target_parameters": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "lora_bias": "none",
+  "lora_dtype": null,
+  "lorap_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "lora_ga_batch_size": 2,
+  "lora_ga_iters": 2,
+  "lora_ga_max_length": 1024,
+  "lora_ga_direction": "ArB2r",
+  "lora_ga_scale": "stable",
+  "lora_ga_stable_gamma": 16,
+  "init_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "use_galore": false,
+  "galore_target_modules": null,
+  "galore_rank": 128,
+  "galore_update_proj_gap": 50,
+  "galore_scale": 1.0,
+  "galore_proj_type": "std",
+  "galore_optim_per_parameter": false,
+  "galore_with_embedding": false,
+  "galore_quantization": false,
+  "galore_proj_quant": false,
+  "galore_proj_bits": 4,
+  "galore_proj_group_size": 256,
+  "galore_cos_threshold": 0.4,
+  "galore_gamma_proj": 2,
+  "galore_queue_size": 5,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "lisa_activated_layers": 0,
+  "lisa_step_interval": 20,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "swanlab_token": null,
+  "swanlab_project": null,
+  "swanlab_workspace": null,
+  "swanlab_exp_name": null,
+  "swanlab_lark_webhook_url": null,
+  "swanlab_lark_secret": null,
+  "swanlab_mode": "cloud",
+  "add_version": true,
+  "create_checkpoint_symlink": false,
+  "zero_hpz_partition_size": null,
+  "deepspeed_autotp_size": null,
+  "early_stop_interval": null,
+  "rank": -1,
+  "global_world_size": 1,
+  "local_world_size": 1,
+  "model_suffix": "Step-Audio-2-mini",
+  "model_info": "ModelInfo(model_type='step_audio2_mini', model_dir='/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini', torch_dtype=torch.bfloat16, max_model_len=16384, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=True, config=None, task_type='causal_lm', num_labels=None)",
+  "model_meta": "ModelMeta(model_type='step_audio2_mini', model_groups=[ModelGroup(models=[Model(ms_model_id='stepfun-ai/Step-Audio-2-mini', hf_model_id='stepfun-ai/Step-Audio-2-mini', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='step_audio2_mini', get_function=<function get_model_tokenizer_step_audio2_mini at 0x7fdf2b1c04c0>, model_arch=MultiModelKeys(arch_name='step_audio2_mini', embedding=None, module_list=None, lm_head=None, q_proj=None, k_proj=None, v_proj=None, o_proj=None, attention=None, mlp=None, down_proj=None, qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None, language_model=['model', 'lm_head'], aligner=['adapter'], vision_tower=['encoder'], generator=[]), architectures=['StepAudio2ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=True, is_reward=False, is_reranker=False, task_type=None, ignore_patterns=None, requires=['transformers==4.53.3', 'torchaudio', 'librosa'], tags=['audio'])",
+  "model_dir": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "_val_dataset_exists": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "hub": "<class 'swift.hub.hub.MSHub'>",
+  "training_args": "Seq2SeqTrainingArguments(output_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=16, per_device_eval_batch_size=16, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=4, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=5e-06, weight_decay=0.01, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=1.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.02, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=100, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=2000, save_total_limit=40, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=2000, dataloader_num_workers=1, dataloader_prefetch_factor=10, past_index=-1, run_name='step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['wandb'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000', hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs={'use_reentrant': False}, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy='steps', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, eval_use_gather_object=False, average_tokens_across_devices=None, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=1e-05, vit_lr=5e-06, use_logits_to_keep=None, ds3_gather_for_generation=True, resume_only_model=False, optimizer='multimodal', loss_type=None, metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='lora', local_repo_path=None, galore_config=None, padding_side='right', padding_free=False, task_type='causal_lm', problem_type=None)"
+}

v0-20260123-014416/v1-20260123-200453/checkpoint-6000/README.md

@@ -0,0 +1,202 @@
+---
+base_model: /data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.11.1

v0-20260123-014416/v1-20260123-200453/checkpoint-6000/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [],
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "^(encoder.*\\.(query|mlp.0|mlp.2|out|value|key)|adapter.*\\.(linear1|linear2))$",
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

v0-20260123-014416/v1-20260123-200453/checkpoint-6000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50a0e1eda4ba03132030371f0fa7d66bbda63e1cb52788ff97b33a49dad31151
+size 11988912

v0-20260123-014416/v1-20260123-200453/checkpoint-6000/additional_config.json

@@ -0,0 +1 @@
+{"lora_dtype": null, "lorap_lr_ratio": null, "lorap_emb_lr": 1e-06}

v0-20260123-014416/v1-20260123-200453/checkpoint-6000/args.json

@@ -0,0 +1,351 @@
+{
+  "output_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453",
+  "overwrite_output_dir": false,
+  "do_train": false,
+  "do_eval": false,
+  "do_predict": false,
+  "eval_strategy": "steps",
+  "prediction_loss_only": false,
+  "per_device_train_batch_size": 16,
+  "per_device_eval_batch_size": 16,
+  "per_gpu_train_batch_size": null,
+  "per_gpu_eval_batch_size": null,
+  "gradient_accumulation_steps": 4,
+  "eval_accumulation_steps": null,
+  "eval_delay": 0,
+  "torch_empty_cache_steps": null,
+  "learning_rate": 5e-06,
+  "weight_decay": 0.01,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "max_grad_norm": 1.0,
+  "num_train_epochs": 1.0,
+  "max_steps": -1,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": null,
+  "warmup_ratio": 0.02,
+  "warmup_steps": 0,
+  "log_level": "passive",
+  "log_level_replica": "warning",
+  "log_on_each_node": true,
+  "logging_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453/runs",
+  "logging_strategy": "steps",
+  "logging_first_step": true,
+  "logging_steps": 100,
+  "logging_nan_inf_filter": true,
+  "save_strategy": "steps",
+  "save_steps": 2000.0,
+  "save_total_limit": 40,
+  "save_safetensors": true,
+  "save_on_each_node": false,
+  "save_only_model": false,
+  "restore_callback_states_from_checkpoint": false,
+  "no_cuda": false,
+  "use_cpu": false,
+  "use_mps_device": false,
+  "seed": 42,
+  "data_seed": 42,
+  "jit_mode_eval": false,
+  "use_ipex": false,
+  "bf16": true,
+  "fp16": false,
+  "fp16_opt_level": "O1",
+  "half_precision_backend": "auto",
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "local_rank": -1,
+  "ddp_backend": null,
+  "tpu_num_cores": null,
+  "tpu_metrics_debug": false,
+  "debug": null,
+  "dataloader_drop_last": false,
+  "eval_steps": 2000.0,
+  "dataloader_num_workers": null,
+  "dataloader_prefetch_factor": null,
+  "past_index": -1,
+  "run_name": "step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF",
+  "disable_tqdm": null,
+  "remove_unused_columns": true,
+  "label_names": null,
+  "load_best_model_at_end": false,
+  "metric_for_best_model": "loss",
+  "greater_is_better": false,
+  "ignore_data_skip": false,
+  "fsdp": "",
+  "fsdp_min_num_params": 0,
+  "fsdp_config": null,
+  "fsdp_transformer_layer_cls_to_wrap": null,
+  "accelerator_config": {
+    "dispatch_batches": false
+  },
+  "deepspeed": null,
+  "label_smoothing_factor": 0.0,
+  "optim": "adamw_torch",
+  "optim_args": null,
+  "adafactor": false,
+  "group_by_length": false,
+  "length_column_name": "length",
+  "report_to": [
+    "wandb"
+  ],
+  "ddp_find_unused_parameters": null,
+  "ddp_bucket_cap_mb": null,
+  "ddp_broadcast_buffers": null,
+  "dataloader_pin_memory": true,
+  "dataloader_persistent_workers": false,
+  "skip_memory_metrics": true,
+  "use_legacy_prediction_loop": false,
+  "push_to_hub": false,
+  "resume_from_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000",
+  "hub_model_id": null,
+  "hub_strategy": "every_save",
+  "hub_token": null,
+  "hub_private_repo": null,
+  "hub_always_push": false,
+  "gradient_checkpointing": true,
+  "gradient_checkpointing_kwargs": "{\"use_reentrant\": false}",
+  "include_inputs_for_metrics": false,
+  "include_for_metrics": [],
+  "eval_do_concat_batches": true,
+  "fp16_backend": "auto",
+  "evaluation_strategy": "steps",
+  "push_to_hub_model_id": null,
+  "push_to_hub_organization": null,
+  "push_to_hub_token": null,
+  "mp_parameters": "",
+  "auto_find_batch_size": false,
+  "full_determinism": false,
+  "torchdynamo": null,
+  "ray_scope": "last",
+  "ddp_timeout": 18000000,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null,
+  "dispatch_batches": null,
+  "split_batches": null,
+  "include_tokens_per_second": false,
+  "include_num_input_tokens_seen": false,
+  "neftune_noise_alpha": null,
+  "optim_target_modules": null,
+  "batch_eval_metrics": false,
+  "eval_on_start": false,
+  "use_liger_kernel": false,
+  "eval_use_gather_object": false,
+  "average_tokens_across_devices": false,
+  "sortish_sampler": false,
+  "predict_with_generate": false,
+  "generation_max_length": null,
+  "generation_num_beams": null,
+  "generation_config": null,
+  "tuner_backend": "peft",
+  "vit_gradient_checkpointing": null,
+  "router_aux_loss_coef": 0.0,
+  "enable_dft_loss": false,
+  "enable_channel_loss": false,
+  "check_model": true,
+  "acc_strategy": "token",
+  "train_dataloader_shuffle": true,
+  "max_epochs": null,
+  "aligner_lr": 1e-05,
+  "vit_lr": 5e-06,
+  "use_logits_to_keep": null,
+  "ds3_gather_for_generation": true,
+  "resume_only_model": false,
+  "optimizer": null,
+  "loss_type": null,
+  "metric": null,
+  "eval_use_evalscope": false,
+  "eval_dataset": [],
+  "eval_dataset_args": null,
+  "eval_limit": null,
+  "eval_generation_config": null,
+  "extra_eval_args": null,
+  "use_flash_ckpt": false,
+  "use_ray": false,
+  "ray_exp_name": null,
+  "device_groups": null,
+  "model": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "model_type": "step_audio2_mini",
+  "model_revision": null,
+  "task_type": "causal_lm",
+  "torch_dtype": "bfloat16",
+  "attn_impl": null,
+  "new_special_tokens": [],
+  "num_labels": null,
+  "problem_type": null,
+  "rope_scaling": null,
+  "device_map": null,
+  "max_memory": {},
+  "max_model_len": null,
+  "local_repo_path": null,
+  "init_strategy": null,
+  "template": "step_audio2_mini",
+  "system": null,
+  "max_length": 16384,
+  "truncation_strategy": "delete",
+  "max_pixels": null,
+  "agent_template": null,
+  "norm_bbox": null,
+  "use_chat_template": true,
+  "padding_free": false,
+  "padding_side": "right",
+  "loss_scale": "default",
+  "sequence_parallel_size": 1,
+  "response_prefix": null,
+  "template_backend": "swift",
+  "dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_train90.jsonl"
+  ],
+  "val_dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "cached_dataset": [],
+  "cached_val_dataset": [],
+  "split_dataset_ratio": 0.0,
+  "dataset_num_proc": 1,
+  "load_from_cache_file": false,
+  "dataset_shuffle": true,
+  "val_dataset_shuffle": false,
+  "streaming": false,
+  "interleave_prob": null,
+  "stopping_strategy": "first_exhausted",
+  "shuffle_buffer_size": 1000,
+  "download_mode": "reuse_dataset_if_exists",
+  "columns": {},
+  "strict": false,
+  "model_name": null,
+  "model_author": null,
+  "custom_dataset_info": [],
+  "quant_method": null,
+  "quant_bits": null,
+  "hqq_axis": null,
+  "bnb_4bit_compute_dtype": "bfloat16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "max_new_tokens": 64,
+  "temperature": 0.0,
+  "top_k": null,
+  "top_p": null,
+  "repetition_penalty": null,
+  "num_beams": 1,
+  "stream": false,
+  "stop_words": [],
+  "logprobs": false,
+  "top_logprobs": null,
+  "ckpt_dir": null,
+  "lora_modules": [],
+  "train_type": "lora",
+  "adapters": [],
+  "external_plugins": [],
+  "model_kwargs": {},
+  "load_args": false,
+  "load_data_args": false,
+  "packing": false,
+  "packing_length": null,
+  "packing_num_proc": 1,
+  "lazy_tokenize": true,
+  "custom_register_path": [],
+  "use_hf": false,
+  "ignore_args_error": false,
+  "use_swift_lora": false,
+  "freeze_parameters": [],
+  "freeze_parameters_regex": null,
+  "freeze_parameters_ratio": 0.0,
+  "trainable_parameters": [],
+  "trainable_parameters_regex": null,
+  "freeze_llm": true,
+  "freeze_vit": false,
+  "freeze_aligner": false,
+  "target_modules": [
+    "all-linear"
+  ],
+  "target_regex": null,
+  "target_parameters": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "lora_bias": "none",
+  "lora_dtype": null,
+  "lorap_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "lora_ga_batch_size": 2,
+  "lora_ga_iters": 2,
+  "lora_ga_max_length": 1024,
+  "lora_ga_direction": "ArB2r",
+  "lora_ga_scale": "stable",
+  "lora_ga_stable_gamma": 16,
+  "init_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "use_galore": false,
+  "galore_target_modules": null,
+  "galore_rank": 128,
+  "galore_update_proj_gap": 50,
+  "galore_scale": 1.0,
+  "galore_proj_type": "std",
+  "galore_optim_per_parameter": false,
+  "galore_with_embedding": false,
+  "galore_quantization": false,
+  "galore_proj_quant": false,
+  "galore_proj_bits": 4,
+  "galore_proj_group_size": 256,
+  "galore_cos_threshold": 0.4,
+  "galore_gamma_proj": 2,
+  "galore_queue_size": 5,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "lisa_activated_layers": 0,
+  "lisa_step_interval": 20,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "swanlab_token": null,
+  "swanlab_project": null,
+  "swanlab_workspace": null,
+  "swanlab_exp_name": null,
+  "swanlab_lark_webhook_url": null,
+  "swanlab_lark_secret": null,
+  "swanlab_mode": "cloud",
+  "add_version": true,
+  "create_checkpoint_symlink": false,
+  "zero_hpz_partition_size": null,
+  "deepspeed_autotp_size": null,
+  "early_stop_interval": null,
+  "rank": -1,
+  "global_world_size": 1,
+  "local_world_size": 1,
+  "model_suffix": "Step-Audio-2-mini",
+  "model_info": "ModelInfo(model_type='step_audio2_mini', model_dir='/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini', torch_dtype=torch.bfloat16, max_model_len=16384, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=True, config=None, task_type='causal_lm', num_labels=None)",
+  "model_meta": "ModelMeta(model_type='step_audio2_mini', model_groups=[ModelGroup(models=[Model(ms_model_id='stepfun-ai/Step-Audio-2-mini', hf_model_id='stepfun-ai/Step-Audio-2-mini', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='step_audio2_mini', get_function=<function get_model_tokenizer_step_audio2_mini at 0x7fdf2b1c04c0>, model_arch=MultiModelKeys(arch_name='step_audio2_mini', embedding=None, module_list=None, lm_head=None, q_proj=None, k_proj=None, v_proj=None, o_proj=None, attention=None, mlp=None, down_proj=None, qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None, language_model=['model', 'lm_head'], aligner=['adapter'], vision_tower=['encoder'], generator=[]), architectures=['StepAudio2ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=True, is_reward=False, is_reranker=False, task_type=None, ignore_patterns=None, requires=['transformers==4.53.3', 'torchaudio', 'librosa'], tags=['audio'])",
+  "model_dir": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "_val_dataset_exists": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "hub": "<class 'swift.hub.hub.MSHub'>",
+  "training_args": "Seq2SeqTrainingArguments(output_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=16, per_device_eval_batch_size=16, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=4, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=5e-06, weight_decay=0.01, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=1.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.02, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=100, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=2000, save_total_limit=40, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=2000, dataloader_num_workers=1, dataloader_prefetch_factor=10, past_index=-1, run_name='step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['wandb'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000', hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs={'use_reentrant': False}, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy='steps', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, eval_use_gather_object=False, average_tokens_across_devices=None, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=1e-05, vit_lr=5e-06, use_logits_to_keep=None, ds3_gather_for_generation=True, resume_only_model=False, optimizer='multimodal', loss_type=None, metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='lora', local_repo_path=None, galore_config=None, padding_side='right', padding_free=False, task_type='causal_lm', problem_type=None)"
+}

v0-20260123-014416/v1-20260123-200453/checkpoint-6000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc9b96b78c9730edc2d4dc127460baa4f6a329480bcffd1918a9f1a51943afb5
+size 24205999

v0-20260123-014416/v1-20260123-200453/checkpoint-6000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31ab0803353fc2e7510dbdc9e623d0a5732ec637183db5671f27b071bfc5e3bc
+size 14645

v0-20260123-014416/v1-20260123-200453/checkpoint-6000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:753710a5ff55912e7374c1a6e5097327041812bfc690b36bba9fc912adb2d1db
+size 1465

v0-20260123-014416/v1-20260123-200453/checkpoint-6000/trainer_state.json

@@ -0,0 +1,548 @@
+{
+  "best_metric": 0.55598271,
+  "best_model_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000",
+  "epoch": 0.6411626415900834,
+  "eval_steps": 2000,
+  "global_step": 6000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00010686044026501389,
+      "grad_norm": 1.6953125,
+      "learning_rate": 2.6595744680851065e-08,
+      "loss": 0.7167500853538513,
+      "step": 1,
+      "token_acc": 0.8417487684729064
+    },
+    {
+      "epoch": 0.010686044026501388,
+      "grad_norm": 1.4453125,
+      "learning_rate": 2.6595744680851065e-06,
+      "loss": 0.6973034637142913,
+      "step": 100,
+      "token_acc": 0.8467176205711577
+    },
+    {
+      "epoch": 0.021372088053002777,
+      "grad_norm": 1.046875,
+      "learning_rate": 4.999978873236658e-06,
+      "loss": 0.6577812957763672,
+      "step": 200,
+      "token_acc": 0.8531807124137688
+    },
+    {
+      "epoch": 0.03205813207950417,
+      "grad_norm": 1.328125,
+      "learning_rate": 4.998159847367393e-06,
+      "loss": 0.6022152709960937,
+      "step": 300,
+      "token_acc": 0.8620946875218537
+    },
+    {
+      "epoch": 0.04274417610600555,
+      "grad_norm": 1.2421875,
+      "learning_rate": 4.993408991262632e-06,
+      "loss": 0.5575482177734375,
+      "step": 400,
+      "token_acc": 0.8680139909957781
+    },
+    {
+      "epoch": 0.05343022013250694,
+      "grad_norm": 1.8984375,
+      "learning_rate": 4.985731880507775e-06,
+      "loss": 0.5264178466796875,
+      "step": 500,
+      "token_acc": 0.8696029378557931
+    },
+    {
+      "epoch": 0.06411626415900834,
+      "grad_norm": 1.1171875,
+      "learning_rate": 4.975137524928774e-06,
+      "loss": 0.4771532440185547,
+      "step": 600,
+      "token_acc": 0.8748165713816137
+    },
+    {
+      "epoch": 0.07480230818550973,
+      "grad_norm": 1.4765625,
+      "learning_rate": 4.961638358018247e-06,
+      "loss": 0.4643104553222656,
+      "step": 700,
+      "token_acc": 0.8754063671076119
+    },
+    {
+      "epoch": 0.0854883522120111,
+      "grad_norm": 1.109375,
+      "learning_rate": 4.945250222343571e-06,
+      "loss": 0.4561585235595703,
+      "step": 800,
+      "token_acc": 0.8795579017210171
+    },
+    {
+      "epoch": 0.0961743962385125,
+      "grad_norm": 1.171875,
+      "learning_rate": 4.925992350954112e-06,
+      "loss": 0.4534299850463867,
+      "step": 900,
+      "token_acc": 0.8851879828938838
+    },
+    {
+      "epoch": 0.10686044026501389,
+      "grad_norm": 1.171875,
+      "learning_rate": 4.903887344809403e-06,
+      "loss": 0.42084632873535155,
+      "step": 1000,
+      "token_acc": 0.8909899302321495
+    },
+    {
+      "epoch": 0.11754648429151528,
+      "grad_norm": 1.390625,
+      "learning_rate": 4.8789611462547414e-06,
+      "loss": 0.43503849029541014,
+      "step": 1100,
+      "token_acc": 0.8900309663514769
+    },
+    {
+      "epoch": 0.12823252831801668,
+      "grad_norm": 1.21875,
+      "learning_rate": 4.851243008575364e-06,
+      "loss": 0.4078071594238281,
+      "step": 1200,
+      "token_acc": 0.8972086577408744
+    },
+    {
+      "epoch": 0.13891857234451807,
+      "grad_norm": 1.3046875,
+      "learning_rate": 4.8207654616649035e-06,
+      "loss": 0.4169163131713867,
+      "step": 1300,
+      "token_acc": 0.8944712980277317
+    },
+    {
+      "epoch": 0.14960461637101946,
+      "grad_norm": 1.25,
+      "learning_rate": 4.7875642738484475e-06,
+      "loss": 0.41801651000976564,
+      "step": 1400,
+      "token_acc": 0.8954553844138678
+    },
+    {
+      "epoch": 0.16029066039752085,
+      "grad_norm": 1.0390625,
+      "learning_rate": 4.7516784099049705e-06,
+      "loss": 0.42521297454833984,
+      "step": 1500,
+      "token_acc": 0.8927475600806661
+    },
+    {
+      "epoch": 0.1709767044240222,
+      "grad_norm": 1.0625,
+      "learning_rate": 4.713149985338428e-06,
+      "loss": 0.40242053985595705,
+      "step": 1600,
+      "token_acc": 0.8985869393070196
+    },
+    {
+      "epoch": 0.1816627484505236,
+      "grad_norm": 1.6953125,
+      "learning_rate": 4.672024216951176e-06,
+      "loss": 0.39753116607666017,
+      "step": 1700,
+      "token_acc": 0.9003083728795708
+    },
+    {
+      "epoch": 0.192348792477025,
+      "grad_norm": 1.328125,
+      "learning_rate": 4.628349369777712e-06,
+      "loss": 0.40670780181884764,
+      "step": 1800,
+      "token_acc": 0.8985113659223496
+    },
+    {
+      "epoch": 0.20303483650352638,
+      "grad_norm": 1.609375,
+      "learning_rate": 4.58217670044103e-06,
+      "loss": 0.4040363311767578,
+      "step": 1900,
+      "token_acc": 0.898509424415663
+    },
+    {
+      "epoch": 0.21372088053002777,
+      "grad_norm": 1.3203125,
+      "learning_rate": 4.533560396998055e-06,
+      "loss": 0.3897912216186523,
+      "step": 2000,
+      "token_acc": 0.9022499382853919
+    },
+    {
+      "epoch": 0.21372088053002777,
+      "eval_loss": 0.5564050674438477,
+      "eval_runtime": 1628.334,
+      "eval_samples_per_second": 20.433,
+      "eval_steps_per_second": 1.277,
+      "eval_token_acc": 0.8750522224756502,
+      "step": 2000
+    },
+    {
+      "epoch": 0.22440692455652916,
+      "grad_norm": 1.390625,
+      "learning_rate": 4.482557515344757e-06,
+      "loss": 0.39870925903320314,
+      "step": 2100,
+      "token_acc": 0.8990360044410765
+    },
+    {
+      "epoch": 0.23509296858303055,
+      "grad_norm": 1.140625,
+      "learning_rate": 4.429227912255589e-06,
+      "loss": 0.38395835876464846,
+      "step": 2200,
+      "token_acc": 0.902773223403757
+    },
+    {
+      "epoch": 0.24577901260953194,
+      "grad_norm": 0.96484375,
+      "learning_rate": 4.3736341751358166e-06,
+      "loss": 0.39471343994140623,
+      "step": 2300,
+      "token_acc": 0.9010956332106893
+    },
+    {
+      "epoch": 0.25646505663603336,
+      "grad_norm": 1.4140625,
+      "learning_rate": 4.315841548569193e-06,
+      "loss": 0.39796932220458986,
+      "step": 2400,
+      "token_acc": 0.9007656570069128
+    },
+    {
+      "epoch": 0.2671511006625347,
+      "grad_norm": 1.265625,
+      "learning_rate": 4.255917857747183e-06,
+      "loss": 0.3913582992553711,
+      "step": 2500,
+      "token_acc": 0.9015067434468572
+    },
+    {
+      "epoch": 0.27783714468903614,
+      "grad_norm": 1.46875,
+      "learning_rate": 4.193933428869592e-06,
+      "loss": 0.38879886627197263,
+      "step": 2600,
+      "token_acc": 0.9019667027692453
+    },
+    {
+      "epoch": 0.2885231887155375,
+      "grad_norm": 2.4375,
+      "learning_rate": 4.129961006610028e-06,
+      "loss": 0.3972347259521484,
+      "step": 2700,
+      "token_acc": 0.9001422292689183
+    },
+    {
+      "epoch": 0.2992092327420389,
+      "grad_norm": 1.2578125,
+      "learning_rate": 4.064075668743043e-06,
+      "loss": 0.3885000991821289,
+      "step": 2800,
+      "token_acc": 0.9027788505528588
+    },
+    {
+      "epoch": 0.3098952767685403,
+      "grad_norm": 1.265625,
+      "learning_rate": 3.996354738033166e-06,
+      "loss": 0.39520988464355467,
+      "step": 2900,
+      "token_acc": 0.9011279097672186
+    },
+    {
+      "epoch": 0.3205813207950417,
+      "grad_norm": 0.921875,
+      "learning_rate": 3.926877691489222e-06,
+      "loss": 0.3871217346191406,
+      "step": 3000,
+      "token_acc": 0.9022083313787118
+    },
+    {
+      "epoch": 0.33126736482154306,
+      "grad_norm": 1.375,
+      "learning_rate": 3.855726067090433e-06,
+      "loss": 0.3810308074951172,
+      "step": 3100,
+      "token_acc": 0.9034641205430405
+    },
+    {
+      "epoch": 0.3419534088480444,
+      "grad_norm": 1.5546875,
+      "learning_rate": 3.7829833680937804e-06,
+      "loss": 0.37891117095947263,
+      "step": 3200,
+      "token_acc": 0.9025715320536037
+    },
+    {
+      "epoch": 0.35263945287454584,
+      "grad_norm": 1.59375,
+      "learning_rate": 3.7087349650349234e-06,
+      "loss": 0.40724864959716794,
+      "step": 3300,
+      "token_acc": 0.8990014549892049
+    },
+    {
+      "epoch": 0.3633254969010472,
+      "grad_norm": 1.1171875,
+      "learning_rate": 3.6330679955376843e-06,
+      "loss": 0.39273796081542967,
+      "step": 3400,
+      "token_acc": 0.900933421493117
+    },
+    {
+      "epoch": 0.3740115409275486,
+      "grad_norm": 1.5,
+      "learning_rate": 3.556071262049684e-06,
+      "loss": 0.3975730514526367,
+      "step": 3500,
+      "token_acc": 0.8998019157822584
+    },
+    {
+      "epoch": 0.38469758495405,
+      "grad_norm": 2.140625,
+      "learning_rate": 3.477835127624158e-06,
+      "loss": 0.37645999908447264,
+      "step": 3600,
+      "token_acc": 0.9051004274755519
+    },
+    {
+      "epoch": 0.3953836289805514,
+      "grad_norm": 1.1796875,
+      "learning_rate": 3.39845140987024e-06,
+      "loss": 0.4185586166381836,
+      "step": 3700,
+      "token_acc": 0.8963374986700084
+    },
+    {
+      "epoch": 0.40606967300705277,
+      "grad_norm": 1.0390625,
+      "learning_rate": 3.318013273196187e-06,
+      "loss": 0.3819176483154297,
+      "step": 3800,
+      "token_acc": 0.9037970813391503
+    },
+    {
+      "epoch": 0.4167557170335542,
+      "grad_norm": 0.8828125,
+      "learning_rate": 3.2366151194720163e-06,
+      "loss": 0.4014278793334961,
+      "step": 3900,
+      "token_acc": 0.8989897804049096
+    },
+    {
+      "epoch": 0.42744176106005555,
+      "grad_norm": 1.2265625,
+      "learning_rate": 3.1543524772398464e-06,
+      "loss": 0.3903131103515625,
+      "step": 4000,
+      "token_acc": 0.9023779507523624
+    },
+    {
+      "epoch": 0.42744176106005555,
+      "eval_loss": 0.5559827089309692,
+      "eval_runtime": 1625.5851,
+      "eval_samples_per_second": 20.468,
+      "eval_steps_per_second": 1.28,
+      "eval_token_acc": 0.8759747254509199,
+      "step": 4000
+    },
+    {
+      "epoch": 0.43812780508655697,
+      "grad_norm": 1.5078125,
+      "learning_rate": 3.071321889601996e-06,
+      "loss": 0.3918019866943359,
+      "step": 4100,
+      "token_acc": 0.9021251892681234
+    },
+    {
+      "epoch": 0.4488138491130583,
+      "grad_norm": 1.3125,
+      "learning_rate": 2.9876208009183837e-06,
+      "loss": 0.39420204162597655,
+      "step": 4200,
+      "token_acc": 0.9012835119864208
+    },
+    {
+      "epoch": 0.45949989313955975,
+      "grad_norm": 3.140625,
+      "learning_rate": 2.9033474424462336e-06,
+      "loss": 0.39763076782226564,
+      "step": 4300,
+      "token_acc": 0.9007918217810081
+    },
+    {
+      "epoch": 0.4701859371660611,
+      "grad_norm": 1.1328125,
+      "learning_rate": 2.8186007170562607e-06,
+      "loss": 0.3739194869995117,
+      "step": 4400,
+      "token_acc": 0.9055805296059564
+    },
+    {
+      "epoch": 0.4808719811925625,
+      "grad_norm": 1.015625,
+      "learning_rate": 2.733480083160674e-06,
+      "loss": 0.3924733734130859,
+      "step": 4500,
+      "token_acc": 0.9005020191807561
+    },
+    {
+      "epoch": 0.4915580252190639,
+      "grad_norm": 1.140625,
+      "learning_rate": 2.648085437989179e-06,
+      "loss": 0.3818220520019531,
+      "step": 4600,
+      "token_acc": 0.9031686537753115
+    },
+    {
+      "epoch": 0.5022440692455653,
+      "grad_norm": 1.1875,
+      "learning_rate": 2.5625170003500053e-06,
+      "loss": 0.3947904968261719,
+      "step": 4700,
+      "token_acc": 0.9003777792274801
+    },
+    {
+      "epoch": 0.5129301132720667,
+      "grad_norm": 0.98046875,
+      "learning_rate": 2.4768751930135102e-06,
+      "loss": 0.39525905609130857,
+      "step": 4800,
+      "token_acc": 0.9012966596534361
+    },
+    {
+      "epoch": 0.523616157298568,
+      "grad_norm": 0.8984375,
+      "learning_rate": 2.391260524856434e-06,
+      "loss": 0.3915714645385742,
+      "step": 4900,
+      "token_acc": 0.9022641509433962
+    },
+    {
+      "epoch": 0.5343022013250694,
+      "grad_norm": 1.3671875,
+      "learning_rate": 2.3057734729050834e-06,
+      "loss": 0.3926304626464844,
+      "step": 5000,
+      "token_acc": 0.9013136721378624
+    },
+    {
+      "epoch": 0.5449882453515709,
+      "grad_norm": 1.34375,
+      "learning_rate": 2.2205143644159065e-06,
+      "loss": 0.3773085403442383,
+      "step": 5100,
+      "token_acc": 0.9033588372573927
+    },
+    {
+      "epoch": 0.5556742893780723,
+      "grad_norm": 1.109375,
+      "learning_rate": 2.135583259131831e-06,
+      "loss": 0.39842689514160157,
+      "step": 5200,
+      "token_acc": 0.9005042691639996
+    },
+    {
+      "epoch": 0.5663603334045736,
+      "grad_norm": 0.890625,
+      "learning_rate": 2.0510798318525585e-06,
+      "loss": 0.39046684265136716,
+      "step": 5300,
+      "token_acc": 0.9020825721592985
+    },
+    {
+      "epoch": 0.577046377431075,
+      "grad_norm": 0.92578125,
+      "learning_rate": 1.9671032554566205e-06,
+      "loss": 0.39618133544921874,
+      "step": 5400,
+      "token_acc": 0.9000712137530825
+    },
+    {
+      "epoch": 0.5877324214575764,
+      "grad_norm": 0.96484375,
+      "learning_rate": 1.8837520845124985e-06,
+      "loss": 0.3802578353881836,
+      "step": 5500,
+      "token_acc": 0.9039277912871451
+    },
+    {
+      "epoch": 0.5984184654840778,
+      "grad_norm": 2.265625,
+      "learning_rate": 1.8011241396153775e-06,
+      "loss": 0.40635391235351564,
+      "step": 5600,
+      "token_acc": 0.8981563749444691
+    },
+    {
+      "epoch": 0.6091045095105792,
+      "grad_norm": 1.53125,
+      "learning_rate": 1.7193163925852968e-06,
+      "loss": 0.36938583374023437,
+      "step": 5700,
+      "token_acc": 0.9048525545411809
+    },
+    {
+      "epoch": 0.6197905535370806,
+      "grad_norm": 1.0703125,
+      "learning_rate": 1.6384248526614122e-06,
+      "loss": 0.37309234619140624,
+      "step": 5800,
+      "token_acc": 0.9052111091372061
+    },
+    {
+      "epoch": 0.630476597563582,
+      "grad_norm": 1.515625,
+      "learning_rate": 1.558544453825951e-06,
+      "loss": 0.3821083068847656,
+      "step": 5900,
+      "token_acc": 0.9030327643665118
+    },
+    {
+      "epoch": 0.6411626415900834,
+      "grad_norm": 1.0390625,
+      "learning_rate": 1.479768943390068e-06,
+      "loss": 0.3839926910400391,
+      "step": 6000,
+      "token_acc": 0.9032827727270024
+    },
+    {
+      "epoch": 0.6411626415900834,
+      "eval_loss": 0.5562668442726135,
+      "eval_runtime": 1881.2991,
+      "eval_samples_per_second": 17.686,
+      "eval_steps_per_second": 1.106,
+      "eval_token_acc": 0.8759600467010075,
+      "step": 6000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 9358,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 2000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.827843079406936e+18,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

v0-20260123-014416/v1-20260123-200453/checkpoint-6000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:859f9e6349e507974dc4ff76def0df20e49d08364f05bd0551845ae28b369820
+size 7377

v0-20260123-014416/v1-20260123-200453/checkpoint-8000/README.md

@@ -0,0 +1,202 @@
+---
+base_model: /data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.11.1

v0-20260123-014416/v1-20260123-200453/checkpoint-8000/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [],
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "^(encoder.*\\.(query|mlp.0|mlp.2|out|value|key)|adapter.*\\.(linear1|linear2))$",
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

v0-20260123-014416/v1-20260123-200453/checkpoint-8000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbf77831dbe7d86e3688535db09f8cf5423a04343f5e3fb26a5ff6a7456da1e7
+size 11988912

v0-20260123-014416/v1-20260123-200453/checkpoint-8000/additional_config.json

@@ -0,0 +1 @@
+{"lora_dtype": null, "lorap_lr_ratio": null, "lorap_emb_lr": 1e-06}

v0-20260123-014416/v1-20260123-200453/checkpoint-8000/args.json

@@ -0,0 +1,351 @@
+{
+  "output_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453",
+  "overwrite_output_dir": false,
+  "do_train": false,
+  "do_eval": false,
+  "do_predict": false,
+  "eval_strategy": "steps",
+  "prediction_loss_only": false,
+  "per_device_train_batch_size": 16,
+  "per_device_eval_batch_size": 16,
+  "per_gpu_train_batch_size": null,
+  "per_gpu_eval_batch_size": null,
+  "gradient_accumulation_steps": 4,
+  "eval_accumulation_steps": null,
+  "eval_delay": 0,
+  "torch_empty_cache_steps": null,
+  "learning_rate": 5e-06,
+  "weight_decay": 0.01,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "max_grad_norm": 1.0,
+  "num_train_epochs": 1.0,
+  "max_steps": -1,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": null,
+  "warmup_ratio": 0.02,
+  "warmup_steps": 0,
+  "log_level": "passive",
+  "log_level_replica": "warning",
+  "log_on_each_node": true,
+  "logging_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453/runs",
+  "logging_strategy": "steps",
+  "logging_first_step": true,
+  "logging_steps": 100,
+  "logging_nan_inf_filter": true,
+  "save_strategy": "steps",
+  "save_steps": 2000.0,
+  "save_total_limit": 40,
+  "save_safetensors": true,
+  "save_on_each_node": false,
+  "save_only_model": false,
+  "restore_callback_states_from_checkpoint": false,
+  "no_cuda": false,
+  "use_cpu": false,
+  "use_mps_device": false,
+  "seed": 42,
+  "data_seed": 42,
+  "jit_mode_eval": false,
+  "use_ipex": false,
+  "bf16": true,
+  "fp16": false,
+  "fp16_opt_level": "O1",
+  "half_precision_backend": "auto",
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "local_rank": -1,
+  "ddp_backend": null,
+  "tpu_num_cores": null,
+  "tpu_metrics_debug": false,
+  "debug": null,
+  "dataloader_drop_last": false,
+  "eval_steps": 2000.0,
+  "dataloader_num_workers": null,
+  "dataloader_prefetch_factor": null,
+  "past_index": -1,
+  "run_name": "step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF",
+  "disable_tqdm": null,
+  "remove_unused_columns": true,
+  "label_names": null,
+  "load_best_model_at_end": false,
+  "metric_for_best_model": "loss",
+  "greater_is_better": false,
+  "ignore_data_skip": false,
+  "fsdp": "",
+  "fsdp_min_num_params": 0,
+  "fsdp_config": null,
+  "fsdp_transformer_layer_cls_to_wrap": null,
+  "accelerator_config": {
+    "dispatch_batches": false
+  },
+  "deepspeed": null,
+  "label_smoothing_factor": 0.0,
+  "optim": "adamw_torch",
+  "optim_args": null,
+  "adafactor": false,
+  "group_by_length": false,
+  "length_column_name": "length",
+  "report_to": [
+    "wandb"
+  ],
+  "ddp_find_unused_parameters": null,
+  "ddp_bucket_cap_mb": null,
+  "ddp_broadcast_buffers": null,
+  "dataloader_pin_memory": true,
+  "dataloader_persistent_workers": false,
+  "skip_memory_metrics": true,
+  "use_legacy_prediction_loop": false,
+  "push_to_hub": false,
+  "resume_from_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000",
+  "hub_model_id": null,
+  "hub_strategy": "every_save",
+  "hub_token": null,
+  "hub_private_repo": null,
+  "hub_always_push": false,
+  "gradient_checkpointing": true,
+  "gradient_checkpointing_kwargs": "{\"use_reentrant\": false}",
+  "include_inputs_for_metrics": false,
+  "include_for_metrics": [],
+  "eval_do_concat_batches": true,
+  "fp16_backend": "auto",
+  "evaluation_strategy": "steps",
+  "push_to_hub_model_id": null,
+  "push_to_hub_organization": null,
+  "push_to_hub_token": null,
+  "mp_parameters": "",
+  "auto_find_batch_size": false,
+  "full_determinism": false,
+  "torchdynamo": null,
+  "ray_scope": "last",
+  "ddp_timeout": 18000000,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null,
+  "dispatch_batches": null,
+  "split_batches": null,
+  "include_tokens_per_second": false,
+  "include_num_input_tokens_seen": false,
+  "neftune_noise_alpha": null,
+  "optim_target_modules": null,
+  "batch_eval_metrics": false,
+  "eval_on_start": false,
+  "use_liger_kernel": false,
+  "eval_use_gather_object": false,
+  "average_tokens_across_devices": false,
+  "sortish_sampler": false,
+  "predict_with_generate": false,
+  "generation_max_length": null,
+  "generation_num_beams": null,
+  "generation_config": null,
+  "tuner_backend": "peft",
+  "vit_gradient_checkpointing": null,
+  "router_aux_loss_coef": 0.0,
+  "enable_dft_loss": false,
+  "enable_channel_loss": false,
+  "check_model": true,
+  "acc_strategy": "token",
+  "train_dataloader_shuffle": true,
+  "max_epochs": null,
+  "aligner_lr": 1e-05,
+  "vit_lr": 5e-06,
+  "use_logits_to_keep": null,
+  "ds3_gather_for_generation": true,
+  "resume_only_model": false,
+  "optimizer": null,
+  "loss_type": null,
+  "metric": null,
+  "eval_use_evalscope": false,
+  "eval_dataset": [],
+  "eval_dataset_args": null,
+  "eval_limit": null,
+  "eval_generation_config": null,
+  "extra_eval_args": null,
+  "use_flash_ckpt": false,
+  "use_ray": false,
+  "ray_exp_name": null,
+  "device_groups": null,
+  "model": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "model_type": "step_audio2_mini",
+  "model_revision": null,
+  "task_type": "causal_lm",
+  "torch_dtype": "bfloat16",
+  "attn_impl": null,
+  "new_special_tokens": [],
+  "num_labels": null,
+  "problem_type": null,
+  "rope_scaling": null,
+  "device_map": null,
+  "max_memory": {},
+  "max_model_len": null,
+  "local_repo_path": null,
+  "init_strategy": null,
+  "template": "step_audio2_mini",
+  "system": null,
+  "max_length": 16384,
+  "truncation_strategy": "delete",
+  "max_pixels": null,
+  "agent_template": null,
+  "norm_bbox": null,
+  "use_chat_template": true,
+  "padding_free": false,
+  "padding_side": "right",
+  "loss_scale": "default",
+  "sequence_parallel_size": 1,
+  "response_prefix": null,
+  "template_backend": "swift",
+  "dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_train90.jsonl"
+  ],
+  "val_dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "cached_dataset": [],
+  "cached_val_dataset": [],
+  "split_dataset_ratio": 0.0,
+  "dataset_num_proc": 1,
+  "load_from_cache_file": false,
+  "dataset_shuffle": true,
+  "val_dataset_shuffle": false,
+  "streaming": false,
+  "interleave_prob": null,
+  "stopping_strategy": "first_exhausted",
+  "shuffle_buffer_size": 1000,
+  "download_mode": "reuse_dataset_if_exists",
+  "columns": {},
+  "strict": false,
+  "model_name": null,
+  "model_author": null,
+  "custom_dataset_info": [],
+  "quant_method": null,
+  "quant_bits": null,
+  "hqq_axis": null,
+  "bnb_4bit_compute_dtype": "bfloat16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "max_new_tokens": 64,
+  "temperature": 0.0,
+  "top_k": null,
+  "top_p": null,
+  "repetition_penalty": null,
+  "num_beams": 1,
+  "stream": false,
+  "stop_words": [],
+  "logprobs": false,
+  "top_logprobs": null,
+  "ckpt_dir": null,
+  "lora_modules": [],
+  "train_type": "lora",
+  "adapters": [],
+  "external_plugins": [],
+  "model_kwargs": {},
+  "load_args": false,
+  "load_data_args": false,
+  "packing": false,
+  "packing_length": null,
+  "packing_num_proc": 1,
+  "lazy_tokenize": true,
+  "custom_register_path": [],
+  "use_hf": false,
+  "ignore_args_error": false,
+  "use_swift_lora": false,
+  "freeze_parameters": [],
+  "freeze_parameters_regex": null,
+  "freeze_parameters_ratio": 0.0,
+  "trainable_parameters": [],
+  "trainable_parameters_regex": null,
+  "freeze_llm": true,
+  "freeze_vit": false,
+  "freeze_aligner": false,
+  "target_modules": [
+    "all-linear"
+  ],
+  "target_regex": null,
+  "target_parameters": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "lora_bias": "none",
+  "lora_dtype": null,
+  "lorap_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "lora_ga_batch_size": 2,
+  "lora_ga_iters": 2,
+  "lora_ga_max_length": 1024,
+  "lora_ga_direction": "ArB2r",
+  "lora_ga_scale": "stable",
+  "lora_ga_stable_gamma": 16,
+  "init_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "use_galore": false,
+  "galore_target_modules": null,
+  "galore_rank": 128,
+  "galore_update_proj_gap": 50,
+  "galore_scale": 1.0,
+  "galore_proj_type": "std",
+  "galore_optim_per_parameter": false,
+  "galore_with_embedding": false,
+  "galore_quantization": false,
+  "galore_proj_quant": false,
+  "galore_proj_bits": 4,
+  "galore_proj_group_size": 256,
+  "galore_cos_threshold": 0.4,
+  "galore_gamma_proj": 2,
+  "galore_queue_size": 5,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "lisa_activated_layers": 0,
+  "lisa_step_interval": 20,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "swanlab_token": null,
+  "swanlab_project": null,
+  "swanlab_workspace": null,
+  "swanlab_exp_name": null,
+  "swanlab_lark_webhook_url": null,
+  "swanlab_lark_secret": null,
+  "swanlab_mode": "cloud",
+  "add_version": true,
+  "create_checkpoint_symlink": false,
+  "zero_hpz_partition_size": null,
+  "deepspeed_autotp_size": null,
+  "early_stop_interval": null,
+  "rank": -1,
+  "global_world_size": 1,
+  "local_world_size": 1,
+  "model_suffix": "Step-Audio-2-mini",
+  "model_info": "ModelInfo(model_type='step_audio2_mini', model_dir='/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini', torch_dtype=torch.bfloat16, max_model_len=16384, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=True, config=None, task_type='causal_lm', num_labels=None)",
+  "model_meta": "ModelMeta(model_type='step_audio2_mini', model_groups=[ModelGroup(models=[Model(ms_model_id='stepfun-ai/Step-Audio-2-mini', hf_model_id='stepfun-ai/Step-Audio-2-mini', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='step_audio2_mini', get_function=<function get_model_tokenizer_step_audio2_mini at 0x7fdf2b1c04c0>, model_arch=MultiModelKeys(arch_name='step_audio2_mini', embedding=None, module_list=None, lm_head=None, q_proj=None, k_proj=None, v_proj=None, o_proj=None, attention=None, mlp=None, down_proj=None, qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None, language_model=['model', 'lm_head'], aligner=['adapter'], vision_tower=['encoder'], generator=[]), architectures=['StepAudio2ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=True, is_reward=False, is_reranker=False, task_type=None, ignore_patterns=None, requires=['transformers==4.53.3', 'torchaudio', 'librosa'], tags=['audio'])",
+  "model_dir": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "_val_dataset_exists": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "hub": "<class 'swift.hub.hub.MSHub'>",
+  "training_args": "Seq2SeqTrainingArguments(output_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=16, per_device_eval_batch_size=16, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=4, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=5e-06, weight_decay=0.01, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=1.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.02, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=100, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=2000, save_total_limit=40, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=2000, dataloader_num_workers=1, dataloader_prefetch_factor=10, past_index=-1, run_name='step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['wandb'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000', hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs={'use_reentrant': False}, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy='steps', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, eval_use_gather_object=False, average_tokens_across_devices=None, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=1e-05, vit_lr=5e-06, use_logits_to_keep=None, ds3_gather_for_generation=True, resume_only_model=False, optimizer='multimodal', loss_type=None, metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='lora', local_repo_path=None, galore_config=None, padding_side='right', padding_free=False, task_type='causal_lm', problem_type=None)"
+}

v0-20260123-014416/v1-20260123-200453/checkpoint-8000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eaa54ce4d6f12732345e3b95c9ee67831a6193ad74077acda3f2ba955eb265fc
+size 24205999

v0-20260123-014416/v1-20260123-200453/checkpoint-8000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89ffad4d855b6ed34e153b28369a58961a10b7b742b35f86bf7be3f0a44a3466
+size 14645

v0-20260123-014416/v1-20260123-200453/checkpoint-8000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:770c50c069dfed51cddb935a91c450607285f8f9a9c450d76c7abaf90044540f
+size 1465

v0-20260123-014416/v1-20260123-200453/checkpoint-8000/trainer_state.json

@@ -0,0 +1,717 @@
+{
+  "best_metric": 0.55598271,
+  "best_model_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000",
+  "epoch": 0.8548835221201111,
+  "eval_steps": 2000,
+  "global_step": 8000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00010686044026501389,
+      "grad_norm": 1.6953125,
+      "learning_rate": 2.6595744680851065e-08,
+      "loss": 0.7167500853538513,
+      "step": 1,
+      "token_acc": 0.8417487684729064
+    },
+    {
+      "epoch": 0.010686044026501388,
+      "grad_norm": 1.4453125,
+      "learning_rate": 2.6595744680851065e-06,
+      "loss": 0.6973034637142913,
+      "step": 100,
+      "token_acc": 0.8467176205711577
+    },
+    {
+      "epoch": 0.021372088053002777,
+      "grad_norm": 1.046875,
+      "learning_rate": 4.999978873236658e-06,
+      "loss": 0.6577812957763672,
+      "step": 200,
+      "token_acc": 0.8531807124137688
+    },
+    {
+      "epoch": 0.03205813207950417,
+      "grad_norm": 1.328125,
+      "learning_rate": 4.998159847367393e-06,
+      "loss": 0.6022152709960937,
+      "step": 300,
+      "token_acc": 0.8620946875218537
+    },
+    {
+      "epoch": 0.04274417610600555,
+      "grad_norm": 1.2421875,
+      "learning_rate": 4.993408991262632e-06,
+      "loss": 0.5575482177734375,
+      "step": 400,
+      "token_acc": 0.8680139909957781
+    },
+    {
+      "epoch": 0.05343022013250694,
+      "grad_norm": 1.8984375,
+      "learning_rate": 4.985731880507775e-06,
+      "loss": 0.5264178466796875,
+      "step": 500,
+      "token_acc": 0.8696029378557931
+    },
+    {
+      "epoch": 0.06411626415900834,
+      "grad_norm": 1.1171875,
+      "learning_rate": 4.975137524928774e-06,
+      "loss": 0.4771532440185547,
+      "step": 600,
+      "token_acc": 0.8748165713816137
+    },
+    {
+      "epoch": 0.07480230818550973,
+      "grad_norm": 1.4765625,
+      "learning_rate": 4.961638358018247e-06,
+      "loss": 0.4643104553222656,
+      "step": 700,
+      "token_acc": 0.8754063671076119
+    },
+    {
+      "epoch": 0.0854883522120111,
+      "grad_norm": 1.109375,
+      "learning_rate": 4.945250222343571e-06,
+      "loss": 0.4561585235595703,
+      "step": 800,
+      "token_acc": 0.8795579017210171
+    },
+    {
+      "epoch": 0.0961743962385125,
+      "grad_norm": 1.171875,
+      "learning_rate": 4.925992350954112e-06,
+      "loss": 0.4534299850463867,
+      "step": 900,
+      "token_acc": 0.8851879828938838
+    },
+    {
+      "epoch": 0.10686044026501389,
+      "grad_norm": 1.171875,
+      "learning_rate": 4.903887344809403e-06,
+      "loss": 0.42084632873535155,
+      "step": 1000,
+      "token_acc": 0.8909899302321495
+    },
+    {
+      "epoch": 0.11754648429151528,
+      "grad_norm": 1.390625,
+      "learning_rate": 4.8789611462547414e-06,
+      "loss": 0.43503849029541014,
+      "step": 1100,
+      "token_acc": 0.8900309663514769
+    },
+    {
+      "epoch": 0.12823252831801668,
+      "grad_norm": 1.21875,
+      "learning_rate": 4.851243008575364e-06,
+      "loss": 0.4078071594238281,
+      "step": 1200,
+      "token_acc": 0.8972086577408744
+    },
+    {
+      "epoch": 0.13891857234451807,
+      "grad_norm": 1.3046875,
+      "learning_rate": 4.8207654616649035e-06,
+      "loss": 0.4169163131713867,
+      "step": 1300,
+      "token_acc": 0.8944712980277317
+    },
+    {
+      "epoch": 0.14960461637101946,
+      "grad_norm": 1.25,
+      "learning_rate": 4.7875642738484475e-06,
+      "loss": 0.41801651000976564,
+      "step": 1400,
+      "token_acc": 0.8954553844138678
+    },
+    {
+      "epoch": 0.16029066039752085,
+      "grad_norm": 1.0390625,
+      "learning_rate": 4.7516784099049705e-06,
+      "loss": 0.42521297454833984,
+      "step": 1500,
+      "token_acc": 0.8927475600806661
+    },
+    {
+      "epoch": 0.1709767044240222,
+      "grad_norm": 1.0625,
+      "learning_rate": 4.713149985338428e-06,
+      "loss": 0.40242053985595705,
+      "step": 1600,
+      "token_acc": 0.8985869393070196
+    },
+    {
+      "epoch": 0.1816627484505236,
+      "grad_norm": 1.6953125,
+      "learning_rate": 4.672024216951176e-06,
+      "loss": 0.39753116607666017,
+      "step": 1700,
+      "token_acc": 0.9003083728795708
+    },
+    {
+      "epoch": 0.192348792477025,
+      "grad_norm": 1.328125,
+      "learning_rate": 4.628349369777712e-06,
+      "loss": 0.40670780181884764,
+      "step": 1800,
+      "token_acc": 0.8985113659223496
+    },
+    {
+      "epoch": 0.20303483650352638,
+      "grad_norm": 1.609375,
+      "learning_rate": 4.58217670044103e-06,
+      "loss": 0.4040363311767578,
+      "step": 1900,
+      "token_acc": 0.898509424415663
+    },
+    {
+      "epoch": 0.21372088053002777,
+      "grad_norm": 1.3203125,
+      "learning_rate": 4.533560396998055e-06,
+      "loss": 0.3897912216186523,
+      "step": 2000,
+      "token_acc": 0.9022499382853919
+    },
+    {
+      "epoch": 0.21372088053002777,
+      "eval_loss": 0.5564050674438477,
+      "eval_runtime": 1628.334,
+      "eval_samples_per_second": 20.433,
+      "eval_steps_per_second": 1.277,
+      "eval_token_acc": 0.8750522224756502,
+      "step": 2000
+    },
+    {
+      "epoch": 0.22440692455652916,
+      "grad_norm": 1.390625,
+      "learning_rate": 4.482557515344757e-06,
+      "loss": 0.39870925903320314,
+      "step": 2100,
+      "token_acc": 0.8990360044410765
+    },
+    {
+      "epoch": 0.23509296858303055,
+      "grad_norm": 1.140625,
+      "learning_rate": 4.429227912255589e-06,
+      "loss": 0.38395835876464846,
+      "step": 2200,
+      "token_acc": 0.902773223403757
+    },
+    {
+      "epoch": 0.24577901260953194,
+      "grad_norm": 0.96484375,
+      "learning_rate": 4.3736341751358166e-06,
+      "loss": 0.39471343994140623,
+      "step": 2300,
+      "token_acc": 0.9010956332106893
+    },
+    {
+      "epoch": 0.25646505663603336,
+      "grad_norm": 1.4140625,
+      "learning_rate": 4.315841548569193e-06,
+      "loss": 0.39796932220458986,
+      "step": 2400,
+      "token_acc": 0.9007656570069128
+    },
+    {
+      "epoch": 0.2671511006625347,
+      "grad_norm": 1.265625,
+      "learning_rate": 4.255917857747183e-06,
+      "loss": 0.3913582992553711,
+      "step": 2500,
+      "token_acc": 0.9015067434468572
+    },
+    {
+      "epoch": 0.27783714468903614,
+      "grad_norm": 1.46875,
+      "learning_rate": 4.193933428869592e-06,
+      "loss": 0.38879886627197263,
+      "step": 2600,
+      "token_acc": 0.9019667027692453
+    },
+    {
+      "epoch": 0.2885231887155375,
+      "grad_norm": 2.4375,
+      "learning_rate": 4.129961006610028e-06,
+      "loss": 0.3972347259521484,
+      "step": 2700,
+      "token_acc": 0.9001422292689183
+    },
+    {
+      "epoch": 0.2992092327420389,
+      "grad_norm": 1.2578125,
+      "learning_rate": 4.064075668743043e-06,
+      "loss": 0.3885000991821289,
+      "step": 2800,
+      "token_acc": 0.9027788505528588
+    },
+    {
+      "epoch": 0.3098952767685403,
+      "grad_norm": 1.265625,
+      "learning_rate": 3.996354738033166e-06,
+      "loss": 0.39520988464355467,
+      "step": 2900,
+      "token_acc": 0.9011279097672186
+    },
+    {
+      "epoch": 0.3205813207950417,
+      "grad_norm": 0.921875,
+      "learning_rate": 3.926877691489222e-06,
+      "loss": 0.3871217346191406,
+      "step": 3000,
+      "token_acc": 0.9022083313787118
+    },
+    {
+      "epoch": 0.33126736482154306,
+      "grad_norm": 1.375,
+      "learning_rate": 3.855726067090433e-06,
+      "loss": 0.3810308074951172,
+      "step": 3100,
+      "token_acc": 0.9034641205430405
+    },
+    {
+      "epoch": 0.3419534088480444,
+      "grad_norm": 1.5546875,
+      "learning_rate": 3.7829833680937804e-06,
+      "loss": 0.37891117095947263,
+      "step": 3200,
+      "token_acc": 0.9025715320536037
+    },
+    {
+      "epoch": 0.35263945287454584,
+      "grad_norm": 1.59375,
+      "learning_rate": 3.7087349650349234e-06,
+      "loss": 0.40724864959716794,
+      "step": 3300,
+      "token_acc": 0.8990014549892049
+    },
+    {
+      "epoch": 0.3633254969010472,
+      "grad_norm": 1.1171875,
+      "learning_rate": 3.6330679955376843e-06,
+      "loss": 0.39273796081542967,
+      "step": 3400,
+      "token_acc": 0.900933421493117
+    },
+    {
+      "epoch": 0.3740115409275486,
+      "grad_norm": 1.5,
+      "learning_rate": 3.556071262049684e-06,
+      "loss": 0.3975730514526367,
+      "step": 3500,
+      "token_acc": 0.8998019157822584
+    },
+    {
+      "epoch": 0.38469758495405,
+      "grad_norm": 2.140625,
+      "learning_rate": 3.477835127624158e-06,
+      "loss": 0.37645999908447264,
+      "step": 3600,
+      "token_acc": 0.9051004274755519
+    },
+    {
+      "epoch": 0.3953836289805514,
+      "grad_norm": 1.1796875,
+      "learning_rate": 3.39845140987024e-06,
+      "loss": 0.4185586166381836,
+      "step": 3700,
+      "token_acc": 0.8963374986700084
+    },
+    {
+      "epoch": 0.40606967300705277,
+      "grad_norm": 1.0390625,
+      "learning_rate": 3.318013273196187e-06,
+      "loss": 0.3819176483154297,
+      "step": 3800,
+      "token_acc": 0.9037970813391503
+    },
+    {
+      "epoch": 0.4167557170335542,
+      "grad_norm": 0.8828125,
+      "learning_rate": 3.2366151194720163e-06,
+      "loss": 0.4014278793334961,
+      "step": 3900,
+      "token_acc": 0.8989897804049096
+    },
+    {
+      "epoch": 0.42744176106005555,
+      "grad_norm": 1.2265625,
+      "learning_rate": 3.1543524772398464e-06,
+      "loss": 0.3903131103515625,
+      "step": 4000,
+      "token_acc": 0.9023779507523624
+    },
+    {
+      "epoch": 0.42744176106005555,
+      "eval_loss": 0.5559827089309692,
+      "eval_runtime": 1625.5851,
+      "eval_samples_per_second": 20.468,
+      "eval_steps_per_second": 1.28,
+      "eval_token_acc": 0.8759747254509199,
+      "step": 4000
+    },
+    {
+      "epoch": 0.43812780508655697,
+      "grad_norm": 1.5078125,
+      "learning_rate": 3.071321889601996e-06,
+      "loss": 0.3918019866943359,
+      "step": 4100,
+      "token_acc": 0.9021251892681234
+    },
+    {
+      "epoch": 0.4488138491130583,
+      "grad_norm": 1.3125,
+      "learning_rate": 2.9876208009183837e-06,
+      "loss": 0.39420204162597655,
+      "step": 4200,
+      "token_acc": 0.9012835119864208
+    },
+    {
+      "epoch": 0.45949989313955975,
+      "grad_norm": 3.140625,
+      "learning_rate": 2.9033474424462336e-06,
+      "loss": 0.39763076782226564,
+      "step": 4300,
+      "token_acc": 0.9007918217810081
+    },
+    {
+      "epoch": 0.4701859371660611,
+      "grad_norm": 1.1328125,
+      "learning_rate": 2.8186007170562607e-06,
+      "loss": 0.3739194869995117,
+      "step": 4400,
+      "token_acc": 0.9055805296059564
+    },
+    {
+      "epoch": 0.4808719811925625,
+      "grad_norm": 1.015625,
+      "learning_rate": 2.733480083160674e-06,
+      "loss": 0.3924733734130859,
+      "step": 4500,
+      "token_acc": 0.9005020191807561
+    },
+    {
+      "epoch": 0.4915580252190639,
+      "grad_norm": 1.140625,
+      "learning_rate": 2.648085437989179e-06,
+      "loss": 0.3818220520019531,
+      "step": 4600,
+      "token_acc": 0.9031686537753115
+    },
+    {
+      "epoch": 0.5022440692455653,
+      "grad_norm": 1.1875,
+      "learning_rate": 2.5625170003500053e-06,
+      "loss": 0.3947904968261719,
+      "step": 4700,
+      "token_acc": 0.9003777792274801
+    },
+    {
+      "epoch": 0.5129301132720667,
+      "grad_norm": 0.98046875,
+      "learning_rate": 2.4768751930135102e-06,
+      "loss": 0.39525905609130857,
+      "step": 4800,
+      "token_acc": 0.9012966596534361
+    },
+    {
+      "epoch": 0.523616157298568,
+      "grad_norm": 0.8984375,
+      "learning_rate": 2.391260524856434e-06,
+      "loss": 0.3915714645385742,
+      "step": 4900,
+      "token_acc": 0.9022641509433962
+    },
+    {
+      "epoch": 0.5343022013250694,
+      "grad_norm": 1.3671875,
+      "learning_rate": 2.3057734729050834e-06,
+      "loss": 0.3926304626464844,
+      "step": 5000,
+      "token_acc": 0.9013136721378624
+    },
+    {
+      "epoch": 0.5449882453515709,
+      "grad_norm": 1.34375,
+      "learning_rate": 2.2205143644159065e-06,
+      "loss": 0.3773085403442383,
+      "step": 5100,
+      "token_acc": 0.9033588372573927
+    },
+    {
+      "epoch": 0.5556742893780723,
+      "grad_norm": 1.109375,
+      "learning_rate": 2.135583259131831e-06,
+      "loss": 0.39842689514160157,
+      "step": 5200,
+      "token_acc": 0.9005042691639996
+    },
+    {
+      "epoch": 0.5663603334045736,
+      "grad_norm": 0.890625,
+      "learning_rate": 2.0510798318525585e-06,
+      "loss": 0.39046684265136716,
+      "step": 5300,
+      "token_acc": 0.9020825721592985
+    },
+    {
+      "epoch": 0.577046377431075,
+      "grad_norm": 0.92578125,
+      "learning_rate": 1.9671032554566205e-06,
+      "loss": 0.39618133544921874,
+      "step": 5400,
+      "token_acc": 0.9000712137530825
+    },
+    {
+      "epoch": 0.5877324214575764,
+      "grad_norm": 0.96484375,
+      "learning_rate": 1.8837520845124985e-06,
+      "loss": 0.3802578353881836,
+      "step": 5500,
+      "token_acc": 0.9039277912871451
+    },
+    {
+      "epoch": 0.5984184654840778,
+      "grad_norm": 2.265625,
+      "learning_rate": 1.8011241396153775e-06,
+      "loss": 0.40635391235351564,
+      "step": 5600,
+      "token_acc": 0.8981563749444691
+    },
+    {
+      "epoch": 0.6091045095105792,
+      "grad_norm": 1.53125,
+      "learning_rate": 1.7193163925852968e-06,
+      "loss": 0.36938583374023437,
+      "step": 5700,
+      "token_acc": 0.9048525545411809
+    },
+    {
+      "epoch": 0.6197905535370806,
+      "grad_norm": 1.0703125,
+      "learning_rate": 1.6384248526614122e-06,
+      "loss": 0.37309234619140624,
+      "step": 5800,
+      "token_acc": 0.9052111091372061
+    },
+    {
+      "epoch": 0.630476597563582,
+      "grad_norm": 1.515625,
+      "learning_rate": 1.558544453825951e-06,
+      "loss": 0.3821083068847656,
+      "step": 5900,
+      "token_acc": 0.9030327643665118
+    },
+    {
+      "epoch": 0.6411626415900834,
+      "grad_norm": 1.0390625,
+      "learning_rate": 1.479768943390068e-06,
+      "loss": 0.3839926910400391,
+      "step": 6000,
+      "token_acc": 0.9032827727270024
+    },
+    {
+      "epoch": 0.6411626415900834,
+      "eval_loss": 0.5562668442726135,
+      "eval_runtime": 1881.2991,
+      "eval_samples_per_second": 17.686,
+      "eval_steps_per_second": 1.106,
+      "eval_token_acc": 0.8759600467010075,
+      "step": 6000
+    },
+    {
+      "epoch": 0.6518486856165847,
+      "grad_norm": 1.5,
+      "learning_rate": 1.402190771972389e-06,
+      "loss": 0.3836984634399414,
+      "step": 6100,
+      "token_acc": 0.9032206080718131
+    },
+    {
+      "epoch": 0.6625347296430861,
+      "grad_norm": 1.703125,
+      "learning_rate": 1.3259009849993357e-06,
+      "loss": 0.381051025390625,
+      "step": 6200,
+      "token_acc": 0.9036385351259948
+    },
+    {
+      "epoch": 0.6732207736695875,
+      "grad_norm": 1.3125,
+      "learning_rate": 1.2509891158545895e-06,
+      "loss": 0.3822282028198242,
+      "step": 6300,
+      "token_acc": 0.9037642860928088
+    },
+    {
+      "epoch": 0.6839068176960889,
+      "grad_norm": 1.2265625,
+      "learning_rate": 1.1775430808030698e-06,
+      "loss": 0.3947774505615234,
+      "step": 6400,
+      "token_acc": 0.901950086240276
+    },
+    {
+      "epoch": 0.6945928617225903,
+      "grad_norm": 1.453125,
+      "learning_rate": 1.1056490758127697e-06,
+      "loss": 0.3984523391723633,
+      "step": 6500,
+      "token_acc": 0.9009766255095756
+    },
+    {
+      "epoch": 0.7052789057490917,
+      "grad_norm": 1.0625,
+      "learning_rate": 1.0353914753955118e-06,
+      "loss": 0.36463138580322263,
+      "step": 6600,
+      "token_acc": 0.9069917781813267
+    },
+    {
+      "epoch": 0.7159649497755931,
+      "grad_norm": 1.8984375,
+      "learning_rate": 9.668527335853722e-07,
+      "loss": 0.37465988159179686,
+      "step": 6700,
+      "token_acc": 0.9041866524545509
+    },
+    {
+      "epoch": 0.7266509938020944,
+      "grad_norm": 1.328125,
+      "learning_rate": 9.00113287170953e-07,
+      "loss": 0.3761553955078125,
+      "step": 6800,
+      "token_acc": 0.9049414331104472
+    },
+    {
+      "epoch": 0.7373370378285958,
+      "grad_norm": 1.2578125,
+      "learning_rate": 8.352514612951018e-07,
+      "loss": 0.3891106414794922,
+      "step": 6900,
+      "token_acc": 0.9018820430264891
+    },
+    {
+      "epoch": 0.7480230818550972,
+      "grad_norm": 0.92578125,
+      "learning_rate": 7.723433775328385e-07,
+      "loss": 0.3889284896850586,
+      "step": 7000,
+      "token_acc": 0.9016702188583
+    },
+    {
+      "epoch": 0.7587091258815987,
+      "grad_norm": 1.0703125,
+      "learning_rate": 7.114628645553845e-07,
+      "loss": 0.39446201324462893,
+      "step": 7100,
+      "token_acc": 0.9018489401695255
+    },
+    {
+      "epoch": 0.7693951699081,
+      "grad_norm": 1.0703125,
+      "learning_rate": 6.526813714851355e-07,
+      "loss": 0.38760986328125,
+      "step": 7200,
+      "token_acc": 0.902246221833607
+    },
+    {
+      "epoch": 0.7800812139346014,
+      "grad_norm": 1.0625,
+      "learning_rate": 5.960678840432629e-07,
+      "loss": 0.38959774017333987,
+      "step": 7300,
+      "token_acc": 0.9020091543079612
+    },
+    {
+      "epoch": 0.7907672579611028,
+      "grad_norm": 1.1640625,
+      "learning_rate": 5.416888435883516e-07,
+      "loss": 0.3816986083984375,
+      "step": 7400,
+      "token_acc": 0.9042744583449611
+    },
+    {
+      "epoch": 0.8014533019876042,
+      "grad_norm": 1.7265625,
+      "learning_rate": 4.896080691410895e-07,
+      "loss": 0.3822678756713867,
+      "step": 7500,
+      "token_acc": 0.9037521445862418
+    },
+    {
+      "epoch": 0.8121393460141055,
+      "grad_norm": 1.3125,
+      "learning_rate": 4.398866824865264e-07,
+      "loss": 0.3938971710205078,
+      "step": 7600,
+      "token_acc": 0.9006727321568306
+    },
+    {
+      "epoch": 0.822825390040607,
+      "grad_norm": 1.3359375,
+      "learning_rate": 3.9258303644179734e-07,
+      "loss": 0.3785970687866211,
+      "step": 7700,
+      "token_acc": 0.9045622587379151
+    },
+    {
+      "epoch": 0.8335114340671084,
+      "grad_norm": 1.421875,
+      "learning_rate": 3.4775264637349065e-07,
+      "loss": 0.3850682067871094,
+      "step": 7800,
+      "token_acc": 0.9020322018638262
+    },
+    {
+      "epoch": 0.8441974780936098,
+      "grad_norm": 1.0546875,
+      "learning_rate": 3.0544812504504114e-07,
+      "loss": 0.36783409118652344,
+      "step": 7900,
+      "token_acc": 0.9066081400907046
+    },
+    {
+      "epoch": 0.8548835221201111,
+      "grad_norm": 1.421875,
+      "learning_rate": 2.657191208706031e-07,
+      "loss": 0.39583698272705076,
+      "step": 8000,
+      "token_acc": 0.900744985000499
+    },
+    {
+      "epoch": 0.8548835221201111,
+      "eval_loss": 0.5561216473579407,
+      "eval_runtime": 1878.2163,
+      "eval_samples_per_second": 17.715,
+      "eval_steps_per_second": 1.107,
+      "eval_token_acc": 0.8760808641041333,
+      "step": 8000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 9358,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 2000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.436080223138406e+18,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

v0-20260123-014416/v1-20260123-200453/checkpoint-8000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:859f9e6349e507974dc4ff76def0df20e49d08364f05bd0551845ae28b369820
+size 7377

v0-20260123-014416/v1-20260123-200453/checkpoint-9358/README.md

@@ -0,0 +1,202 @@
+---
+base_model: /data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.11.1

v0-20260123-014416/v1-20260123-200453/checkpoint-9358/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [],
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "^(encoder.*\\.(query|mlp.0|mlp.2|out|value|key)|adapter.*\\.(linear1|linear2))$",
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

v0-20260123-014416/v1-20260123-200453/checkpoint-9358/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba70fd4ed51bb98671b7c6fd2d272db13f2f5c89bd3341778e92e85c67fb6bad
+size 11988912

v0-20260123-014416/v1-20260123-200453/checkpoint-9358/additional_config.json

@@ -0,0 +1 @@
+{"lora_dtype": null, "lorap_lr_ratio": null, "lorap_emb_lr": 1e-06}

v0-20260123-014416/v1-20260123-200453/checkpoint-9358/args.json

@@ -0,0 +1,351 @@
+{
+  "output_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453",
+  "overwrite_output_dir": false,
+  "do_train": false,
+  "do_eval": false,
+  "do_predict": false,
+  "eval_strategy": "steps",
+  "prediction_loss_only": false,
+  "per_device_train_batch_size": 16,
+  "per_device_eval_batch_size": 16,
+  "per_gpu_train_batch_size": null,
+  "per_gpu_eval_batch_size": null,
+  "gradient_accumulation_steps": 4,
+  "eval_accumulation_steps": null,
+  "eval_delay": 0,
+  "torch_empty_cache_steps": null,
+  "learning_rate": 5e-06,
+  "weight_decay": 0.01,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "max_grad_norm": 1.0,
+  "num_train_epochs": 1.0,
+  "max_steps": -1,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": null,
+  "warmup_ratio": 0.02,
+  "warmup_steps": 0,
+  "log_level": "passive",
+  "log_level_replica": "warning",
+  "log_on_each_node": true,
+  "logging_dir": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453/runs",
+  "logging_strategy": "steps",
+  "logging_first_step": true,
+  "logging_steps": 100,
+  "logging_nan_inf_filter": true,
+  "save_strategy": "steps",
+  "save_steps": 2000.0,
+  "save_total_limit": 40,
+  "save_safetensors": true,
+  "save_on_each_node": false,
+  "save_only_model": false,
+  "restore_callback_states_from_checkpoint": false,
+  "no_cuda": false,
+  "use_cpu": false,
+  "use_mps_device": false,
+  "seed": 42,
+  "data_seed": 42,
+  "jit_mode_eval": false,
+  "use_ipex": false,
+  "bf16": true,
+  "fp16": false,
+  "fp16_opt_level": "O1",
+  "half_precision_backend": "auto",
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "local_rank": -1,
+  "ddp_backend": null,
+  "tpu_num_cores": null,
+  "tpu_metrics_debug": false,
+  "debug": null,
+  "dataloader_drop_last": false,
+  "eval_steps": 2000.0,
+  "dataloader_num_workers": null,
+  "dataloader_prefetch_factor": null,
+  "past_index": -1,
+  "run_name": "step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF",
+  "disable_tqdm": null,
+  "remove_unused_columns": true,
+  "label_names": null,
+  "load_best_model_at_end": false,
+  "metric_for_best_model": "loss",
+  "greater_is_better": false,
+  "ignore_data_skip": false,
+  "fsdp": "",
+  "fsdp_min_num_params": 0,
+  "fsdp_config": null,
+  "fsdp_transformer_layer_cls_to_wrap": null,
+  "accelerator_config": {
+    "dispatch_batches": false
+  },
+  "deepspeed": null,
+  "label_smoothing_factor": 0.0,
+  "optim": "adamw_torch",
+  "optim_args": null,
+  "adafactor": false,
+  "group_by_length": false,
+  "length_column_name": "length",
+  "report_to": [
+    "wandb"
+  ],
+  "ddp_find_unused_parameters": null,
+  "ddp_bucket_cap_mb": null,
+  "ddp_broadcast_buffers": null,
+  "dataloader_pin_memory": true,
+  "dataloader_persistent_workers": false,
+  "skip_memory_metrics": true,
+  "use_legacy_prediction_loop": false,
+  "push_to_hub": false,
+  "resume_from_checkpoint": "/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000",
+  "hub_model_id": null,
+  "hub_strategy": "every_save",
+  "hub_token": null,
+  "hub_private_repo": null,
+  "hub_always_push": false,
+  "gradient_checkpointing": true,
+  "gradient_checkpointing_kwargs": "{\"use_reentrant\": false}",
+  "include_inputs_for_metrics": false,
+  "include_for_metrics": [],
+  "eval_do_concat_batches": true,
+  "fp16_backend": "auto",
+  "evaluation_strategy": "steps",
+  "push_to_hub_model_id": null,
+  "push_to_hub_organization": null,
+  "push_to_hub_token": null,
+  "mp_parameters": "",
+  "auto_find_batch_size": false,
+  "full_determinism": false,
+  "torchdynamo": null,
+  "ray_scope": "last",
+  "ddp_timeout": 18000000,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null,
+  "dispatch_batches": null,
+  "split_batches": null,
+  "include_tokens_per_second": false,
+  "include_num_input_tokens_seen": false,
+  "neftune_noise_alpha": null,
+  "optim_target_modules": null,
+  "batch_eval_metrics": false,
+  "eval_on_start": false,
+  "use_liger_kernel": false,
+  "eval_use_gather_object": false,
+  "average_tokens_across_devices": false,
+  "sortish_sampler": false,
+  "predict_with_generate": false,
+  "generation_max_length": null,
+  "generation_num_beams": null,
+  "generation_config": null,
+  "tuner_backend": "peft",
+  "vit_gradient_checkpointing": null,
+  "router_aux_loss_coef": 0.0,
+  "enable_dft_loss": false,
+  "enable_channel_loss": false,
+  "check_model": true,
+  "acc_strategy": "token",
+  "train_dataloader_shuffle": true,
+  "max_epochs": null,
+  "aligner_lr": 1e-05,
+  "vit_lr": 5e-06,
+  "use_logits_to_keep": null,
+  "ds3_gather_for_generation": true,
+  "resume_only_model": false,
+  "optimizer": null,
+  "loss_type": null,
+  "metric": null,
+  "eval_use_evalscope": false,
+  "eval_dataset": [],
+  "eval_dataset_args": null,
+  "eval_limit": null,
+  "eval_generation_config": null,
+  "extra_eval_args": null,
+  "use_flash_ckpt": false,
+  "use_ray": false,
+  "ray_exp_name": null,
+  "device_groups": null,
+  "model": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "model_type": "step_audio2_mini",
+  "model_revision": null,
+  "task_type": "causal_lm",
+  "torch_dtype": "bfloat16",
+  "attn_impl": null,
+  "new_special_tokens": [],
+  "num_labels": null,
+  "problem_type": null,
+  "rope_scaling": null,
+  "device_map": null,
+  "max_memory": {},
+  "max_model_len": null,
+  "local_repo_path": null,
+  "init_strategy": null,
+  "template": "step_audio2_mini",
+  "system": null,
+  "max_length": 16384,
+  "truncation_strategy": "delete",
+  "max_pixels": null,
+  "agent_template": null,
+  "norm_bbox": null,
+  "use_chat_template": true,
+  "padding_free": false,
+  "padding_side": "right",
+  "loss_scale": "default",
+  "sequence_parallel_size": 1,
+  "response_prefix": null,
+  "template_backend": "swift",
+  "dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_train90.jsonl"
+  ],
+  "val_dataset": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "cached_dataset": [],
+  "cached_val_dataset": [],
+  "split_dataset_ratio": 0.0,
+  "dataset_num_proc": 1,
+  "load_from_cache_file": false,
+  "dataset_shuffle": true,
+  "val_dataset_shuffle": false,
+  "streaming": false,
+  "interleave_prob": null,
+  "stopping_strategy": "first_exhausted",
+  "shuffle_buffer_size": 1000,
+  "download_mode": "reuse_dataset_if_exists",
+  "columns": {},
+  "strict": false,
+  "model_name": null,
+  "model_author": null,
+  "custom_dataset_info": [],
+  "quant_method": null,
+  "quant_bits": null,
+  "hqq_axis": null,
+  "bnb_4bit_compute_dtype": "bfloat16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "max_new_tokens": 64,
+  "temperature": 0.0,
+  "top_k": null,
+  "top_p": null,
+  "repetition_penalty": null,
+  "num_beams": 1,
+  "stream": false,
+  "stop_words": [],
+  "logprobs": false,
+  "top_logprobs": null,
+  "ckpt_dir": null,
+  "lora_modules": [],
+  "train_type": "lora",
+  "adapters": [],
+  "external_plugins": [],
+  "model_kwargs": {},
+  "load_args": false,
+  "load_data_args": false,
+  "packing": false,
+  "packing_length": null,
+  "packing_num_proc": 1,
+  "lazy_tokenize": true,
+  "custom_register_path": [],
+  "use_hf": false,
+  "ignore_args_error": false,
+  "use_swift_lora": false,
+  "freeze_parameters": [],
+  "freeze_parameters_regex": null,
+  "freeze_parameters_ratio": 0.0,
+  "trainable_parameters": [],
+  "trainable_parameters_regex": null,
+  "freeze_llm": true,
+  "freeze_vit": false,
+  "freeze_aligner": false,
+  "target_modules": [
+    "all-linear"
+  ],
+  "target_regex": null,
+  "target_parameters": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "lora_bias": "none",
+  "lora_dtype": null,
+  "lorap_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "lora_ga_batch_size": 2,
+  "lora_ga_iters": 2,
+  "lora_ga_max_length": 1024,
+  "lora_ga_direction": "ArB2r",
+  "lora_ga_scale": "stable",
+  "lora_ga_stable_gamma": 16,
+  "init_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "use_galore": false,
+  "galore_target_modules": null,
+  "galore_rank": 128,
+  "galore_update_proj_gap": 50,
+  "galore_scale": 1.0,
+  "galore_proj_type": "std",
+  "galore_optim_per_parameter": false,
+  "galore_with_embedding": false,
+  "galore_quantization": false,
+  "galore_proj_quant": false,
+  "galore_proj_bits": 4,
+  "galore_proj_group_size": 256,
+  "galore_cos_threshold": 0.4,
+  "galore_gamma_proj": 2,
+  "galore_queue_size": 5,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "lisa_activated_layers": 0,
+  "lisa_step_interval": 20,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "swanlab_token": null,
+  "swanlab_project": null,
+  "swanlab_workspace": null,
+  "swanlab_exp_name": null,
+  "swanlab_lark_webhook_url": null,
+  "swanlab_lark_secret": null,
+  "swanlab_mode": "cloud",
+  "add_version": true,
+  "create_checkpoint_symlink": false,
+  "zero_hpz_partition_size": null,
+  "deepspeed_autotp_size": null,
+  "early_stop_interval": null,
+  "rank": -1,
+  "global_world_size": 1,
+  "local_world_size": 1,
+  "model_suffix": "Step-Audio-2-mini",
+  "model_info": "ModelInfo(model_type='step_audio2_mini', model_dir='/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini', torch_dtype=torch.bfloat16, max_model_len=16384, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=True, config=None, task_type='causal_lm', num_labels=None)",
+  "model_meta": "ModelMeta(model_type='step_audio2_mini', model_groups=[ModelGroup(models=[Model(ms_model_id='stepfun-ai/Step-Audio-2-mini', hf_model_id='stepfun-ai/Step-Audio-2-mini', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='step_audio2_mini', get_function=<function get_model_tokenizer_step_audio2_mini at 0x7fdf2b1c04c0>, model_arch=MultiModelKeys(arch_name='step_audio2_mini', embedding=None, module_list=None, lm_head=None, q_proj=None, k_proj=None, v_proj=None, o_proj=None, attention=None, mlp=None, down_proj=None, qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None, language_model=['model', 'lm_head'], aligner=['adapter'], vision_tower=['encoder'], generator=[]), architectures=['StepAudio2ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=True, is_reward=False, is_reranker=False, task_type=None, ignore_patterns=None, requires=['transformers==4.53.3', 'torchaudio', 'librosa'], tags=['audio'])",
+  "model_dir": "/data/pangkaiyu/dg/Step-Audio2/Step-Audio-2-mini",
+  "_val_dataset_exists": [
+    "/data/pangkaiyu/dg/make_jsonl_py_file/lora_merged_0122_55+5+5+1_signal_val5.jsonl"
+  ],
+  "hub": "<class 'swift.hub.hub.MSHub'>",
+  "training_args": "Seq2SeqTrainingArguments(output_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=16, per_device_eval_batch_size=16, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=4, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=5e-06, weight_decay=0.01, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=1.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.02, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/v1-20260123-200453/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=100, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=2000, save_total_limit=40, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=2000, dataloader_num_workers=1, dataloader_prefetch_factor=10, past_index=-1, run_name='step-audio2-mini-enc+align-whole0122_signal_new_dpdc_lora-1gpu-bs16*4_1_gckF', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['wandb'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint='/data/pangkaiyu/dg/output_step_audio2_mini-encoder+align-whole0122_signal_new1_dpdc-lora-1gpu-bs16*4_1_gckF/v0-20260123-014416/checkpoint-4000', hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs={'use_reentrant': False}, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy='steps', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, eval_use_gather_object=False, average_tokens_across_devices=None, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=1e-05, vit_lr=5e-06, use_logits_to_keep=None, ds3_gather_for_generation=True, resume_only_model=False, optimizer='multimodal', loss_type=None, metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='lora', local_repo_path=None, galore_config=None, padding_side='right', padding_free=False, task_type='causal_lm', problem_type=None)"
+}