See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: elyza/Llama-3-ELYZA-JP-8B
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 43e5277b38c5b9ca_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/43e5277b38c5b9ca_train_data.json
  type:
    field_instruction: question
    field_output: answer
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
gradient_clipping: 1.0
group_by_length: false
hub_model_id: auxyus/5a7bf840-7e3f-48a0-8dc7-d36d0b59bb5c
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 3
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_steps: 100
micro_batch_size: 8
mlflow_experiment_name: /tmp/43e5277b38c5b9ca_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 4
sequence_len: 1024
special_tokens:
  pad_token: <|eot_id|>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: techspear-hub
wandb_mode: online
wandb_name: bdc5d8b7-c291-4772-a9f7-28e2904ac8b5
wandb_project: Gradients-On-Two
wandb_run: your_name
wandb_runid: bdc5d8b7-c291-4772-a9f7-28e2904ac8b5
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

5a7bf840-7e3f-48a0-8dc7-d36d0b59bb5c

This model is a fine-tuned version of elyza/Llama-3-ELYZA-JP-8B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.8048

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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: 10
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0068	1	2.4070
2.2245	0.0608	9	2.1583
2.0257	0.1216	18	1.9933
2.0234	0.1824	27	1.9224
1.9812	0.2432	36	1.8780
1.8843	0.3041	45	1.8488
1.8643	0.3649	54	1.8309
1.8706	0.4257	63	1.8197
1.6933	0.4865	72	1.8112
1.7838	0.5473	81	1.8070
1.8249	0.6081	90	1.8045
1.8606	0.6689	99	1.8048

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1