54 lines
1.1 KiB
YAML
54 lines
1.1 KiB
YAML
base_model: openlm-research/open_llama_3b_v2
|
|
# optionally might have model_type or tokenizer_type
|
|
model_type: LlamaForCausalLM
|
|
tokenizer_type: LlamaTokenizer
|
|
# Automatically upload checkpoint and final model to HF
|
|
# hub_model_id: username/custom_model_name
|
|
|
|
load_in_8bit: false
|
|
load_in_4bit: true
|
|
push_dataset_to_hub:
|
|
datasets:
|
|
- path: teknium/GPT4-LLM-Cleaned
|
|
type: alpaca
|
|
dataset_prepared_path:
|
|
val_set_size: 0.05
|
|
adapter: qlora
|
|
lora_model_dir:
|
|
sequence_len: 1024
|
|
sample_packing: true
|
|
lora_r: 8
|
|
lora_alpha: 32
|
|
lora_dropout: 0.05
|
|
lora_target_linear: true
|
|
wandb_project:
|
|
wandb_entity:
|
|
wandb_watch:
|
|
wandb_name:
|
|
wandb_log_model:
|
|
output_dir: ./outputs/qlora-out
|
|
gradient_accumulation_steps: 1
|
|
micro_batch_size: 2
|
|
num_epochs: 4
|
|
optimizer: paged_adamw_32bit
|
|
torchdistx_path:
|
|
lr_scheduler: cosine
|
|
learning_rate: 0.0002
|
|
bf16: false
|
|
fp16: true
|
|
tf32: false
|
|
gradient_checkpointing: true
|
|
resume_from_checkpoint:
|
|
logging_steps: 1
|
|
flash_attention: true
|
|
gptq_groupsize:
|
|
gptq_model_v1:
|
|
warmup_steps: 20
|
|
evals_per_epoch: 4
|
|
saves_per_epoch: 1
|
|
weight_decay: 0.1
|
|
special_tokens:
|
|
bos_token: "<s>"
|
|
eos_token: "</s>"
|
|
unk_token: "<unk>"
|