See axolotl config

axolotl version: 0.6.0

adapter: lora
auto_find_batch_size: true
base_model: aisingapore/llama3-8b-cpt-sea-lionv2.1-instruct
bf16: auto
chat_template: llama3
dataloader_num_workers: 12
dataset_prepared_path: null
datasets:
- data_files:
  - 6c11f7ee9fd5bc58_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/6c11f7ee9fd5bc58_train_data.json
  type:
    field_instruction: prompt
    field_output: chosen_response
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 3
early_stopping_threshold: 0.001
eval_batch_size: 32
eval_max_new_tokens: 128
eval_steps: 500
eval_table_size: null
evals_per_epoch: null
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: false
group_by_length: false
hub_model_id: jssky/3333a13d-a24f-4a98-b13f-a97324b77310
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0003
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: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_steps: 1000
micro_batch_size: 128
mlflow_experiment_name: /tmp/6c11f7ee9fd5bc58_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 5
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
s2_attention: null
sample_packing: false
save_steps: 500
saves_per_epoch: null
sequence_len: 1024
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: 00004a65-1430-449c-be79-5ca6fffeac7e
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 00004a65-1430-449c-be79-5ca6fffeac7e
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

3333a13d-a24f-4a98-b13f-a97324b77310

This model is a fine-tuned version of aisingapore/llama3-8b-cpt-sea-lionv2.1-instruct on the None dataset.

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: 0.0003
• train_batch_size: 128
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 256
• optimizer: Use adamw_bnb_8bit 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: 138

Training results

Framework versions

• PEFT 0.14.0
• Transformers 4.46.3
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3