JUNHO YEOM comments

Repositories
Issues
Comments

Results 3 comments of


                                            JUNHO YEOM

m3 embedding model pos score(similarity) is getting lower

@ngothanhnam0910 Hi, I'm using data not normalized. bellow this comment, I received an answer that I should use data that is not normalized. https://github.com/FlagOpen/FlagEmbedding/issues/701

m3 embedding model pos score(similarity) is getting lower

@staoxiao Hi, Thanks for your comment. In order to remove noise from the data, we use data with a score of 2 or more in the reranker as positive data....

finetuning failing

In my case, In my case, it happened when I didn't apply the "deepspeed" setting. If I didn't apply the `--deepspeed` setting, I shouldn't have applied the `--gradient_checkpointing` option either.