See axolotl config

axolotl version: 0.11.0.dev0

adapter: lora
base_model: oopsung/llama2-7b-koNqa-test-v1
bf16: auto
chat_template: llama3
dataloader_num_workers: 1
dataloader_pin_memory: false
dataset_prepared_path: null
datasets:
- data_files:
  - d573960c-4c46-4d22-9469-55c8c20ddf00_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: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_batch_size: 2
eval_max_new_tokens: 128
eval_sample_packing: false
eval_steps: 41
eval_table_size: null
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: true
greater_is_better: false
group_by_length: false
learning_rate: 0.0002
load_best_model_at_end: true
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 128
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_steps: 82
metric_for_best_model: eval_loss
micro_batch_size: 8
mlflow_experiment_name: /workspace/axolotl/data/d573960c-4c46-4d22-9469-55c8c20ddf00_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: /app/checkpoints/d573960c-4c46-4d22-9469-55c8c20ddf00/god-d573960c-cal
pad_to_sequence_len: false
resume_from_checkpoint: null
s2_attention: null
sample_packing: true
save_steps: 41
save_total_limit: 3
sdp_attention: true
sequence_len: 4096
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.005633802816901409
wandb_entity: null
wandb_mode: offline
wandb_name: d573960c-4c46-4d22-9469-55c8c20ddf00_god-d573960c-cal
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: d573960c-4c46-4d22-9469-55c8c20ddf00_god-d573960c-cal
warmup_ratio: 0.03
weight_decay: 0.01
xformers_attention: null

app/checkpoints/d573960c-4c46-4d22-9469-55c8c20ddf00/god-d573960c-cal

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.3357

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: 0.0002
• train_batch_size: 8
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_BNB 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: 2
• training_steps: 82

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0	0	1.8332
2.2122	0.0203	41	1.4023
1.715	0.0406	82	1.3357

Framework versions

• PEFT 0.15.2
• Transformers 4.53.1
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.4