gpt2-swahili / run.sh
alokmatta
gpt2 swahili
49a1166
raw
history blame
623 Bytes
./run_clm_flax.py \
--output_dir="./" \
--model_type="gpt2" \
--config_name="./" \
--tokenizer_name="./" \
--dataset_name="flax-community/swahili-safi" \
--do_train --do_eval \
--block_size="512" \
--per_device_train_batch_size="64" \
--per_device_eval_batch_size="64" \
--learning_rate="5e-3" --warmup_steps="1000" \
--adam_beta1="0.9" --adam_beta2="0.98" --weight_decay="0.01" \
--overwrite_output_dir \
--num_train_epochs="20" \
--logging_steps="500" \
--save_steps="2500" \
--dtype float32 \
--eval_steps="2500" \
--preprocessing_num_workers="96"