axolotl version: 0.4.0

base_model: croissantllm/CroissantLLMBase                                                                                                                                                                   
model_type: LlamaForCausalLM                                                                                                                                                                                
tokenizer_type: LlamaTokenizerFast                                                                                                                                                                              
is_llama_derived_model: true                                                                                                                                                                                
                                                                                                                                                                                                            
load_in_8bit: false                                                                                                                                                                                         
load_in_4bit: false                                                                                                                                                                                         
strict: false                                                                                                                                                                                               
                                                                                                                                                                                                            
datasets:                                                                                                                                                                                                   
  - path: manu/mmlu_auxiliary_train_formatted_extra
    split: train                                                                                                                                                               
    type: completion                                                                                                                                                                                        

dataset_prepared_path: last_run_prepared                                                                                                                                                                    
val_set_size: 0.05                                                                                                                                                                                          
output_dir: ./out                                                                                                                                                                                           
                                                                                                                                                                                                            
sequence_len: 2048                                                                                                                                                                                          
sample_packing: false                                                                                                                                                                                       
pad_to_sequence_len: false                                                                                                                                                                                   
                                                                                                                                                                                                            
adapter:                                                                                                                                                                                                    
lora_model_dir:                                                                                                                                                                                             
lora_r:                                                                                                                                                                                                     
lora_alpha:                                                                                                                                                                                                 
lora_dropout:
lora_target_linear:
lora_fan_in_fan_out:

wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 2
micro_batch_size: 24
num_epochs: 1
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
flash_attn_cross_entropy: false
flash_attn_rms_norm: true
flash_attn_fuse_qkv: false
flash_attn_fuse_mlp: true

warmup_steps: 50
evals_per_epoch: 4
eval_table_size:
saves_per_epoch: 1
debug:
deepspeed: #deepspeed_configs/zero2.json # multi-gpu only
weight_decay: 0.1
fsdp:
fsdp_config:
special_tokens: