See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: NousResearch/CodeLlama-7b-hf-flash
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 8a11bca18490447d_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/8a11bca18490447d_train_data.json
  type:
    field_input: entities
    field_instruction: document_type
    field_output: text
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map:
  ? ''
  : 0,1,2,4,5,6,7
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/ff7de68d-f54c-472f-8e69-2bdc9b055cf2
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: 128
lora_dropout: 0.3
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 840
micro_batch_size: 4
mlflow_experiment_name: /tmp/8a11bca18490447d_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
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: </s>
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
use_rslora: true
val_set_size: 0.04
wandb_entity: null
wandb_mode: online
wandb_name: a4f08651-7a69-43a5-bc33-f69c995f5d9e
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: a4f08651-7a69-43a5-bc33-f69c995f5d9e
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

ff7de68d-f54c-472f-8e69-2bdc9b055cf2

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

• Loss: 0.3440

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
• distributed_type: multi-GPU
• num_devices: 6
• gradient_accumulation_steps: 8
• total_train_batch_size: 192
• total_eval_batch_size: 24
• 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: 597

Training results

Training Loss	Epoch	Step	Validation Loss
8.1168	0.0034	1	1.0268
3.035	0.3350	100	0.4032
2.9305	0.6700	200	0.3717
2.7215	1.0050	300	0.3549
2.3288	1.3400	400	0.3498
2.2021	1.6750	500	0.3440

Framework versions

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