See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: DeepMount00/Llama-3-8b-Ita
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - b853391e03784fa0_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/b853391e03784fa0_train_data.json
  type:
    field_instruction: content
    field_output: title
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
device_map:
  ? ''
  : 0,1,2,3,4,5,6,7
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: true
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/098c2c40-6f4f-4b17-8b10-0a2e2579c5dd
hub_repo: null
hub_strategy: null
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: 128
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lora_target_modules:
- q_proj
- k_proj
- v_proj
- o_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 903
micro_batch_size: 4
mlflow_experiment_name: /tmp/b853391e03784fa0_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 10
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
sequence_len: 2048
special_tokens:
  pad_token: <|eot_id|>
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.04792026068621813
wandb_entity: null
wandb_mode: online
wandb_name: c8fca9aa-1807-456b-8ada-33a5d696fe9f
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: c8fca9aa-1807-456b-8ada-33a5d696fe9f
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

098c2c40-6f4f-4b17-8b10-0a2e2579c5dd

This model is a fine-tuned version of DeepMount00/Llama-3-8b-Ita on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8104

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: 42
• gradient_accumulation_steps: 8
• 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: 903

Training results

Training Loss	Epoch	Step	Validation Loss
3.2807	0.0003	1	3.3028
1.1652	0.0322	100	1.0695
0.8804	0.0644	200	1.0240
1.1392	0.0966	300	0.9703
0.6268	0.1289	400	0.9438
1.1619	0.1611	500	0.8929
0.5958	0.1933	600	0.8560
0.8048	0.2255	700	0.8280
0.8476	0.2577	800	0.8134
0.6939	0.2899	900	0.8104

Framework versions

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