See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: unsloth/llama-3-8b
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 3387218bff6889ea_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/3387218bff6889ea_train_data.json
  type:
    field_instruction: context
    field_output: response
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 5
eval_max_new_tokens: 128
eval_steps: 50
eval_table_size: null
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 16
gradient_checkpointing: true
group_by_length: false
hub_model_id: sn56t0/54f6f9b4-7250-4b95-9e46-8a4d9a26f5e9
hub_repo: null
hub_strategy: checkpoint
hub_token: null
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: 32
lora_dropout: 0.05
lora_fan_in_fan_out: true
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lora_target_modules:
- q_proj
- k_proj
- v_proj
lr_scheduler: cosine
max_steps: 2384
micro_batch_size: 4
mlflow_experiment_name: /tmp/3387218bff6889ea_train_data.json
model_type: AutoModelForCausalLM
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
seed: 3152046077
sequence_len: 2048
shuffle: true
strict: false
tf32: true
tokenizer_type: AutoTokenizer
torch_compile: true
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: sn56-miner
wandb_mode: disabled
wandb_name: null
wandb_project: god
wandb_run: 2kcs
wandb_runid: null
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

54f6f9b4-7250-4b95-9e46-8a4d9a26f5e9

This model is a fine-tuned version of unsloth/llama-3-8b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.9744

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: 4
• eval_batch_size: 4
• seed: 3152046077
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 16
• total_train_batch_size: 256
• total_eval_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: 10
• training_steps: 615

Training results

Training Loss	Epoch	Step	Validation Loss
3.0246	0.0016	1	3.0606
2.0825	0.0814	50	2.0478
2.0536	0.1628	100	2.0270
2.0386	0.2442	150	2.0189
2.0418	0.3257	200	2.0076
2.0195	0.4071	250	2.0002
2.0159	0.4885	300	1.9939
2.0077	0.5699	350	1.9878
1.9377	0.6513	400	1.9830
1.9941	0.7327	450	1.9789
2.0044	0.8142	500	1.9761
1.936	0.8956	550	1.9747
1.9673	0.9770	600	1.9744

Framework versions

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