22 lines
503 B
YAML
22 lines
503 B
YAML
|
model_name: "EleutherAI/pythia-410m-deduped"
|
||
|
|
output_model_name: "sophiex/pythia-410m-sft_hh_rlhf"
|
||
|
|
dataset_name: "sophiex/hh-rlhf"
|
||
|
|
dataset_text_field: null
|
||
|
|
report_to: "wandb"
|
||
|
|
learning_rate: 1e-5
|
||
|
|
lr_scheduler_type: cosine
|
||
|
|
warmup_steps: 100
|
||
|
|
weight_decay: 0.05
|
||
|
|
load_in_8bit: False
|
||
|
|
fp16: True
|
||
|
|
bf16: False
|
||
|
|
gradient_checkpointing: False
|
||
|
|
gradient_accumulation_steps: 4
|
||
|
|
per_device_train_batch_size: 4
|
||
|
|
per_device_eval_batch_size: 8
|
||
|
|
num_train_epochs: 3
|
||
|
|
max_seq_length: 1024
|
||
|
|
use_peft: True
|
||
|
|
lora_r: 8
|
||
|
|
lora_alpha: 32
|