diff --git a/scripts/gemma_data_exp/gemma_self_gen_qa_and_pretrain.sh b/scripts/gemma_data_exp/gemma_self_gen_qa_and_pretrain.sh index 85589d3..f8319e8 100644 --- a/scripts/gemma_data_exp/gemma_self_gen_qa_and_pretrain.sh +++ b/scripts/gemma_data_exp/gemma_self_gen_qa_and_pretrain.sh @@ -7,7 +7,7 @@ #SBATCH --error=outputs/%x-%j.out uv run accelerate launch --num_processes=4 --gradient_accumulation_steps=32 --gradient_clipping=1.0 \ ---gpu_ids all --main_process_port 29563 intx_sft.py configs/self_gen_3_small_and_pretrain.yaml \ +--gpu_ids all --main_process_port 29563 intx_sft.py configs/self_gen_3_and_pretrain_small.yaml \ --model_name_or_path=google/gemma-2-2b-it --num_train_epochs=2 --per_device_train_batch_size=8 \ --gradient_accumulation_steps=32 --per_device_eval_batch_size=8 --exp_setup=hyper_lora --aggregator_type=perceiver \ --target_modules=down_proj \