See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: NousResearch/Meta-Llama-3-8B
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 77ba6d73fd94c922_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/77ba6d73fd94c922_train_data.json
  type:
    field_instruction: instruction
    field_output: output
    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: 2
flash_attention: false
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: true
hub_model_id: luckjsg/fb623698-1171-4f16-b312-34e7e1621495
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 10
lora_alpha: 64
lora_bias: none
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 128
lora_target_linear: true
lora_target_modules:
- q_proj
- k_proj
- v_proj
- o_proj
- gate_proj
- up_proj
- down_proj
lr_scheduler: cosine
max_steps: 100
micro_batch_size: 2
mlflow_experiment_name: /tmp/77ba6d73fd94c922_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: false
sample_packing: false
saves_per_epoch: 2
sequence_len: 512
special_tokens:
  pad_token: <|end_of_text|>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 0655973e-32fe-419c-b1dd-ea3a9cf3a099
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 0655973e-32fe-419c-b1dd-ea3a9cf3a099
warmup_steps: 50
weight_decay: 0.01
xformers_attention: false

fb623698-1171-4f16-b312-34e7e1621495

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

• Loss: 0.7086

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• 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: 50
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0006	1	1.5952
0.7627	0.0099	17	1.5819
0.7747	0.0198	34	0.8150
0.9702	0.0298	51	0.7884
0.7416	0.0397	68	0.7375
0.6675	0.0496	85	0.7086

Framework versions

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