See axolotl config

axolotl version: 0.10.0.dev0

adapter: null
base_model: /cache/models/JackFram--llama-160m
bf16: true
chat_template: llama3
cosine_min_lr_ratio: 0.3
dataset_prepared_path: null
datasets:
- data_files:
  - /workspace/axolotl/data/f7288a9f-b1ee-4bd0-9a70-4da4d19fcfdd_train_data.json
  ds_type: json
  format: custom
  path: /workspace/axolotl/data
  type:
    field_instruction: instruct
    field_output: output
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
ddp: true
debug: null
deepspeed: null
early_stopping_patience: null
eval_sample_packing: false
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: skrd3/6b482ee4-d135-48d5-8126-e11a38734246
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
liger_fused_linear_cross_entropy: true
liger_glu_activation: true
liger_layer_norm: true
liger_rms_norm: true
liger_rope: true
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 512
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 128
lora_target_linear: true
lr_scheduler: cosine
micro_batch_size: 72
mlflow_experiment_name: /workspace/axolotl/data/f7288a9f-b1ee-4bd0-9a70-4da4d19fcfdd_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: paged_adamw_8bit
output_dir: /app/checkpoints/f7288a9f-b1ee-4bd0-9a70-4da4d19fcfdd/temp_trainer
pad_to_sequence_len: true
plugins:
- axolotl.integrations.liger.LigerPlugin
resume_from_checkpoint: null
s2_attention: null
sample_packing: true
save_only_model: true
saves_per_epoch: 1
sequence_len: 1024
special_tokens:
  pad_token: </s>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
use_liger: true
val_set_size: 0.0011073706591070162
wandb_entity: null
wandb_mode: online
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: default
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

6b482ee4-d135-48d5-8126-e11a38734246

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.9424

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 72
• eval_batch_size: 72
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 288
• optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
2.9946	0.0040	1	2.9424
3.0311	0.9990	248	2.9424
3.0497	1.3625	339	2.9424
3.0508	1.9950	496	2.9424
3.018	2.9909	744	2.9424

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1