Skywork/eval
Liang Zhao 520daef18f
Fix attention bug in eval_loss.py and make it support ChatGLM3 (#65)
* update url

* update skywork tech report arxiv url

* update evaluation data to hugginface and fix some typos

* fix loss typo

* update wise model, add  FQA in loss evaluation

* add faq in evaulation

* fix typo

* update

* fix attention bug in eval_loss.py

---------

Co-authored-by: liang.zhao <liang.zhao@singularity-ai.com>
2023-12-09 18:45:44 +08:00
..
EVALUATION.md init skywork repo 2023-10-30 03:13:04 +00:00
eval_gsm8k.py refine eval script (#4) 2023-10-30 12:10:04 +08:00
eval_loss.py Fix attention bug in eval_loss.py and make it support ChatGLM3 (#65) 2023-12-09 18:45:44 +08:00
evaluate_ceval.py refine eval script (#4) 2023-10-30 12:10:04 +08:00
evaluate_cmmlu.py refine eval script (#4) 2023-10-30 12:10:04 +08:00
evaluate_mmlu.py refine eval script (#4) 2023-10-30 12:10:04 +08:00
gsm8k_prompt.txt init skywork repo 2023-10-30 03:13:04 +00:00