python run_clm_flax.py \
    --output_dir="./" \
    --model_type="gpt2" \
    --model_name_or_path="." \
    --config_name="./" \
    --tokenizer_name="./" \
    --train_file="/mnt/disks/flaxdisk/vgd/vgd_train.json" \
    --validation_file="/mnt/disks/flaxdisk/vgd/vgd_eval.json" \
    --cache_dir="/mnt/disks/flaxdisk/cache/" \
    --do_train --do_eval \
    --block_size="512" \
    --per_device_train_batch_size="64" \
    --per_device_eval_batch_size="64" \
    --learning_rate="1e-3" \
    --warmup_steps="1000" \
    --adam_beta1="0.9" --adam_beta2="0.98" --weight_decay="0.01" \
    --overwrite_output_dir \
    --num_train_epochs="5" \
    --logging_steps="500" \
    --save_steps="1000" \
    --eval_steps="1000" \
    --push_to_hub