See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: unsloth/Llama-3.2-3B
bf16: auto
chat_template: llama3
cosine_min_lr_ratio: 0.1
data_processes: 16
dataset_prepared_path: null
datasets:
- data_files:
  - e5ed75ab6d05ed8d_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/e5ed75ab6d05ed8d_train_data.json
  type:
    field_instruction: text_description
    field_output: text
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: '{'''':torch.cuda.current_device()}'
do_eval: true
early_stopping_patience: 1
eval_batch_size: 3
eval_sample_packing: false
eval_steps: 25
evaluation_strategy: steps
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 10
gradient_checkpointing: true
group_by_length: true
hub_model_id: sn56d1/4c1c9e6f-d82f-471b-90a7-158f9a0bcfd4
hub_repo: stevemonite
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lora_target_modules:
- q_proj
- v_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_memory:
  0: 70GiB
max_steps: 333
micro_batch_size: 3
mlflow_experiment_name: /tmp/e5ed75ab6d05ed8d_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.95
  adam_epsilon: 1e-5
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 50
save_strategy: steps
sequence_len: 2048
strict: false
tf32: false
tokenizer_type: AutoTokenizer
torch_compile: false
train_on_inputs: false
trust_remote_code: true
val_set_size: 50
wandb_entity: sn56-miner
wandb_mode: disabled
wandb_name: null
wandb_project: god
wandb_run: zjkv
wandb_runid: null
warmup_raio: 0.03
warmup_ratio: 0.04
weight_decay: 0.01
xformers_attention: null

4c1c9e6f-d82f-471b-90a7-158f9a0bcfd4

This model is a fine-tuned version of unsloth/Llama-3.2-3B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.1426

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: 3
• eval_batch_size: 3
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 10
• total_train_batch_size: 120
• total_eval_batch_size: 12
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 13
• training_steps: 333

Training results

Training Loss	Epoch	Step	Validation Loss
4.3889	0.0086	1	4.5809
3.9813	0.2146	25	3.5360
3.8343	0.4292	50	3.3974
3.6724	0.6438	75	3.2902
3.7497	0.8584	100	3.2946
3.0159	1.0730	125	3.2453
3.11	1.2876	150	3.2199
3.1524	1.5021	175	3.2106
2.9865	1.7167	200	3.1638
2.9999	1.9313	225	3.1539
2.8174	2.1459	250	3.1418
2.792	2.3605	275	3.1384
2.7043	2.5751	300	3.1426

Framework versions

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