base_model: meta-llama/Llama-2-7b-hf
base_model_config: meta-llama/Llama-2-7b-hf
model_type: MistralForCausalLM
tokenizer_type: LlamaTokenizer
is_mistral_derived_model: true

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: timdettmers/openassistant-guanaco
    type: completion
dataset_prepared_path:
val_set_size: 0.1
output_dir: ./out

sequence_len: 4096
sample_packing: false
pad_to_sequence_len: false

wandb_project: llama-neft-guanaco
wandb_entity:
wandb_watch:
wandb_run_id:
wandb_log_model:

gradient_accumulation_steps: 2
micro_batch_size: 8
num_epochs: 3
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 0.000038

train_on_inputs: false
group_by_length: false
bf16: true
fp16: false
tf32: false

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

noisy_embedding_alpha: 5

warmup_steps: 10
eval_steps: 0.0666666666
save_steps:
debug:
deepspeed: deepspeed/zero2.json
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
  bos_token: "<s>"
  eos_token: "</s>"
  unk_token: "<unk>"