View difference between Paste ID: WT68UaBs and UiQFMQGm

 #!/bin/bash
 # Copyright (c) Microsoft Corporation.
 # SPDX-License-Identifier: Apache-2.0
 
 # DeepSpeed Team
 OUTPUT=$1
 ZERO_STAGE=$2
-if [ "$OUTPUT" == "" ]; then
+
-    OUTPUT=/nobackup/jirayu/models/opt-125m-sft
+
 fi
 mkdir -p $OUTPUT
 
 deepspeed main.py \
    --data_path stanfordnlp/SHP \
    --data_split 10,0,0 \
    --model_name_or_path facebook/opt-125m \
    --per_device_train_batch_size 8 \
    --per_device_eval_batch_size 8 \
    --max_seq_len 512 \
    --learning_rate 1e-5 \
    --weight_decay 0. \
    --num_train_epochs 2 \
    --gradient_accumulation_steps 1 \
    --lr_scheduler_type cosine \
    --num_warmup_steps 0 \
    --seed 1234 \
    --zero_stage $ZERO_STAGE \
    --deepspeed \
    --enable_tensorboard \
    --tensorboard_path $OUTPUT \
    --output_dir $OUTPUT \
    | tee $OUTPUT/training.log