apunkt/doc-to-lora
1
0
Fork 0
mirror of https://github.com/SakanaAI/doc-to-lora.git synced 2026-07-23 17:01:04 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

gemma data exp scripts

Browse source
This commit is contained in:
51616 2025-06-10 15:40:19 +00:00
parent bae2593cd4
commit 4a2960cd64
3 changed files with 29 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

6
scripts/gemma_data_exp/gemma_self_gen_qa.sh
View file

@ -6,10 +6,10 @@
#SBATCH --output=outputs/%x-%j.out
#SBATCH --error=outputs/%x-%j.out
uv run accelerate launch --num_processes=4 --gradient_accumulation_steps=8 --gradient_clipping=1.0 \
uv run accelerate launch --num_processes=4 --gradient_accumulation_steps=16 --gradient_clipping=1.0 \
--gpu_ids all --main_process_port 29562 intx_sft.py configs/self_gen_3_small.yaml \
--model_name_or_path=google/gemma-2-2b-it --num_train_epochs=4 --per_device_train_batch_size=32 \
--gradient_accumulation_steps=8 --per_device_eval_batch_size=32 --exp_setup=hyper_lora --aggregator_type=perceiver \
--model_name_or_path=google/gemma-2-2b-it --num_train_epochs=3 --per_device_train_batch_size=8 \
--gradient_accumulation_steps=16 --per_device_eval_batch_size=4 --exp_setup=hyper_lora --aggregator_type=perceiver \
--target_modules=down_proj \
--num_self_attends_per_block=4 --num_latent_factor=1 \
--lora_r=8 \

4
scripts/gemma_data_exp/gemma_self_gen_qa_and_pretrain.sh
View file

@ -8,8 +8,8 @@
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_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 \
--model_name_or_path=google/gemma-2-2b-it --num_train_epochs=2 --per_device_train_batch_size=4 \
--gradient_accumulation_steps=32 --per_device_eval_batch_size=4 --exp_setup=hyper_lora --aggregator_type=perceiver \
--target_modules=down_proj \
--num_self_attends_per_block=4 --num_latent_factor=1 \
--lora_r=8 \

24
scripts/gemma_data_exp/gemma_self_gen_qa_mini.sh Normal file
View file

@ -0,0 +1,24 @@
#!/bin/bash
#SBATCH --job-name=ctxlora_medium
#SBATCH --partition=a3
#SBATCH --nodes=1
#SBATCH --gpus=4
#SBATCH --output=outputs/%x-%j.out
#SBATCH --error=outputs/%x-%j.out
uv run accelerate launch --num_processes=4 --gradient_accumulation_steps=16 --gradient_clipping=1.0 \
--gpu_ids all --main_process_port 29562 intx_sft.py configs/self_gen_3_mini.yaml \
--model_name_or_path=google/gemma-2-2b-it --num_train_epochs=3 --per_device_train_batch_size=8 \
--gradient_accumulation_steps=16 --per_device_eval_batch_size=4 --exp_setup=hyper_lora --aggregator_type=perceiver \
--target_modules=down_proj \
--num_self_attends_per_block=8 --num_latent_factor=1 \
--lora_r=8 \
--eval_steps=5000 --save_steps=5000 --learning_rate=4e-5 --lora_dropout=0.0 \
--neftune_noise_alpha=5 --use_light_weight_lora=False \
--add_negative_prompt=False \
--add_repeat_prompt=False \
--use_sequence_packing=True --max_packed_inp_len=16384 --max_packed_ctx_len=32768 \
--per_rank_gen=True \
--per_layer_processing=True \
--gen_lora_l1_reg_coef=0.1 \