diff --git a/examples/legacy/llama3/tune_llama3_8b_ptd_full.sh b/examples/legacy/llama3/tune_llama3_8b_ptd_full.sh index 034fa49fe014f39ca4aa7f6856d8e3ac53f8ee89..0d2d7c6da4533343ca96d8422cc25df404675fea 100644 --- a/examples/legacy/llama3/tune_llama3_8b_ptd_full.sh +++ b/examples/legacy/llama3/tune_llama3_8b_ptd_full.sh @@ -98,4 +98,4 @@ torchrun $DISTRIBUTED_ARGS posttrain_gpt.py \ $OUTPUT_ARGS \ --distributed-backend nccl \ --load ${CKPT_LOAD_DIR} \ - | tee logs/train_llama3_8b_full.log + | tee logs/tune_llama3_8b_full.log diff --git a/examples/mcore/deepseek_r1_distill_qwen/tune_distill_qwen_32b_full.sh b/examples/mcore/deepseek_r1_distill_qwen/tune_distill_qwen_32b_full.sh index 241ab0e525fe0c0fdae3a7ef93e7060bc06deb31..9de296b4377d3066cc4f46b480ac3fa7f59d8fdb 100644 --- a/examples/mcore/deepseek_r1_distill_qwen/tune_distill_qwen_32b_full.sh +++ b/examples/mcore/deepseek_r1_distill_qwen/tune_distill_qwen_32b_full.sh @@ -117,4 +117,4 @@ torchrun $DISTRIBUTED_ARGS posttrain_gpt.py \ --load ${CKPT_LOAD_DIR} \ --save ${CKPT_SAVE_DIR} \ --distributed-backend nccl \ - | tee logs/pretrain_mcore_distill_qwen_32b_full_8k.log + | tee logs/tune_mcore_distill_qwen_32b_full_8k.log diff --git a/examples/mcore/llama3/tune_llama3_8b_full_ptd.sh b/examples/mcore/llama3/tune_llama3_8b_full_ptd.sh index 645fd84dcd99673421e616e68888b860670d10f8..d2c1f8fd70bb241a72ae505509f7c6d6e2dda699 100644 --- a/examples/mcore/llama3/tune_llama3_8b_full_ptd.sh +++ b/examples/mcore/llama3/tune_llama3_8b_full_ptd.sh @@ -106,4 +106,4 @@ torchrun $DISTRIBUTED_ARGS posttrain_gpt.py \ $DATA_ARGS \ $OUTPUT_ARGS \ --distributed-backend nccl \ - | tee logs/tune_llama31_8b_full_ptd.log \ No newline at end of file + | tee logs/tune_llama3_8b_full_ptd.log \ No newline at end of file diff --git a/examples/mcore/qwen2/pretrain_qwen2_7b_32k_ptd.sh b/examples/mcore/qwen2/pretrain_qwen2_7b_32k_ptd.sh index 7f9ff4a040cbd2145152693305d30b618b48be3b..31311b1db249ec8bc3969bf62145b4d11ea32839 100644 --- a/examples/mcore/qwen2/pretrain_qwen2_7b_32k_ptd.sh +++ b/examples/mcore/qwen2/pretrain_qwen2_7b_32k_ptd.sh @@ -104,4 +104,4 @@ torchrun $DISTRIBUTED_ARGS pretrain_gpt.py \ --load ${CKPT_LOAD_DIR} \ --save ${CKPT_SAVE_DIR} \ --distributed-backend nccl \ - | tee logs/pretrain_mcore_qwen2_72b_32k.log + | tee logs/pretrain_mcore_qwen2_7b_32k.log