| architecture: |
| training_mode: lora |
| gradient_checkpointing: true |
| intermediate_dropout: 0.0 |
| backbone_kwargs: '{}' |
| dataset: |
| data_sample_choice: |
| - Train |
| - Validation |
| data_sample: 0.2 |
| system_column: null |
| input_column: prompt |
| output_column: answer |
| train_dataset_id: e3a038d6-c009-4935-97a3-d35720bee1de |
| train_dataframe: http://minio:9000/datasets/e3a038d6-c009-4935-97a3-d35720bee1de/e3a038d6-c009-4935-97a3-d35720bee1de?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=minioadmin%2F20251017%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20251017T161637Z&X-Amz-Expires=604800&X-Amz-SignedHeaders=host&X-Amz-Signature=181921582822dc2618a04e8dc3a9978aa6d26de465d977b024400311fb77376d |
| train_dataframe_type: parquet |
| validation_dataframe: null |
| validation_dataframe_type: null |
| validation_strategy: automatic |
| validation_size: 0.2 |
| num_classes: null |
| validation_dataset_id: null |
| unroll_conversations: false |
| environment: |
| find_unused_parameters: false |
| huggingface_branch: main |
| mixed_precision: false |
| use_fsdp: false |
| use_fsdp_cpu_offload: false |
| seed: -1 |
| trust_remote_code: true |
| tokenizer: |
| max_length: 2304 |
| padding_quantile: 1.0 |
| tokenizer_kwargs: '{"use_fast": true, "add_prefix_space": false}' |
| padding_side: left |
| chat_template: null |
| training: |
| attention_implementation: auto |
| batch_size: 1 |
| differential_learning_rate_layers: [] |
| differential_learning_rate: 1.0e-05 |
| epochs: 1 |
| evaluate_before_training: true |
| evaluation_epochs: 1.0 |
| grad_accumulation: 1 |
| gradient_clip: 0.0 |
| learning_rate: 0.0001 |
| lora_alpha: 16 |
| lora_dropout: 0.05 |
| lora_rank: 4 |
| lora_target_modules: '' |
| loss_function: CrossEntropyLoss |
| min_learning_rate_ratio: 0.0 |
| optimizer: AdamW |
| schedule: Cosine |
| train_validation_data: false |
| use_length_based_sampler: false |
| warmup_epochs: 0.0 |
| weight_decay: 0.0 |
| llm_judge_model: '' |
| llm_judge_prompt_template: '' |
| logging: {} |
| prediction: |
| batch_size_inference: 0 |
| max_length_inference: 256 |
| metrics: |
| - BLEU |
| min_length_inference: 2 |
| num_beams: 1 |
| repetition_penalty: 1.0 |
| temperature: 0.0 |
| top_k: 0 |
| top_p: 1.0 |
| experiment_name: hungry-bonobo |
| experiment_id: 74273746-83a7-459b-83ed-c4d06df0d782 |
| project_id: 14673bfa-20ad-486c-815b-f0fed6c6eae8 |
| reference_experiment_ids: [] |
| llm_backbone: h2oai/h2ovl-mississippi-2b |
| problem_type: multimodal_causal_language_modeling |
|
|
