kyy/llm_trainer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
2,236 Commits 1 Branch 0 Tags
843b5d85e98b312e5d41ce62ec10e199011beb8c
Commit Graph
1 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
d 6979f3f848 经过大量的增量预训练，进行对比试验，发现这个bug：llama3在预训练时使用的tokenizer.eos_toke是'<|end_of_text|>' ，这里在每条数据后面也得用这个，而不是'<|eot_id|>'，否则很容易导致严重的性能下降 2024-06-11 16:23:40 +08:00