base_model: furiosa-ai/mlperf-gpt-j-6b
batch_size: 32
bf16: true
chat_template: tokenizer_default_fallback_alpaca
datasets:
- data_files:
  - 80b3f2b5f3ce3209_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/80b3f2b5f3ce3209_train_data.json
  type:
    field_input: headline_a
    field_instruction: rendered_input
    field_output: headline_b
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
eval_steps: 20
flash_attention: true
gpu_memory_limit: 80GiB
gradient_checkpointing: true
group_by_length: true
hub_model_id: SystemAdmin123/5e183a09-3809-4b42-97f6-0e567d3b687b
hub_strategy: checkpoint
learning_rate: 0.0002
logging_steps: 10
lr_scheduler: cosine
micro_batch_size: 3
model_type: AutoModelForCausalLM
num_epochs: 10
optimizer: adamw_bnb_8bit
output_dir: /workspace/axolotl/configs
pad_to_sequence_len: true
resize_token_embeddings_to_32x: false
sample_packing: false
save_steps: 40
save_total_limit: 1
sequence_len: 2048
tokenizer_type: GPT2TokenizerFast
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.1
wandb_entity: ''
wandb_mode: online
wandb_name: furiosa-ai/mlperf-gpt-j-6b-/tmp/80b3f2b5f3ce3209_train_data.json
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: default
warmup_ratio: 0.05
xformers_attention: true