See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: codellama/CodeLlama-7b-hf
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 4b225deccd7d9717_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/4b225deccd7d9717_train_data.json
  type:
    field_input: ds_i
    field_instruction: dg_i
    field_output: dg_o
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device: cuda
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: false
fp16: null
fsdp: null
fsdp_config:
  max_steps: 50
  weight_decay: 0.01
gradient_accumulation_steps: 4
gradient_checkpointing: false
group_by_length: false
hub_model_id: tarabukinivan/6945a446-1172-4349-be07-2060801e3101
hub_repo: null
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: 16
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 8
lora_target_linear: true
lr_scheduler: cosine
max_memory:
  0: 70GiB
max_steps: 50
micro_batch_size: 2
mlflow_experiment_name: /tmp/4b225deccd7d9717_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
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: 70
sequence_len: 2048
special_tokens:
  pad_token: </s>
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: 6945a446-1172-4349-be07-2060801e3101
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 6945a446-1172-4349-be07-2060801e3101
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

6945a446-1172-4349-be07-2060801e3101

This model is a fine-tuned version of codellama/CodeLlama-7b-hf on the None dataset. It achieves the following results on the evaluation set:

• Loss: nan

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_TORCH 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: 50

Training results

Training Loss	Epoch	Step	Validation Loss
0.0	0.0001	1	nan
0.0	0.0003	5	nan
0.0	0.0007	10	nan
0.0	0.0010	15	nan
0.0	0.0013	20	nan
0.0	0.0016	25	nan
0.0	0.0020	30	nan
0.0	0.0023	35	nan
0.0	0.0026	40	nan
0.0	0.0029	45	nan
0.0	0.0033	50	nan

Framework versions

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