haoranxu/Llama-3-Instruct-8B-SimPO
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: haoranxu/Llama-3-Instruct-8B-SimPO
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-05-10 17:37:59 +08:00
commit ac80c37ff2
17 changed files with 413887 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
all_results.json Normal file
View File

@@ -0,0 +1,22 @@
{
"epoch": 1.0,
"eval_logits/chosen": 0.19371525943279266,
"eval_logits/rejected": 0.24551750719547272,
"eval_logps/chosen": -0.5973814725875854,
"eval_logps/rejected": -0.7979374527931213,
"eval_loss": 1.46084463596344,
"eval_rewards/accuracies": 0.6290322542190552,
"eval_rewards/chosen": -1.4934533834457397,
"eval_rewards/margins": 0.5013901591300964,
"eval_rewards/rejected": -1.9948437213897705,
"eval_runtime": 100.7391,
"eval_samples": 1961,
"eval_samples_per_second": 19.466,
"eval_steps_per_second": 0.308,
"total_flos": 0.0,
"train_loss": 1.5168422256779468,
"train_runtime": 9435.3453,
"train_samples": 59876,
"train_samples_per_second": 6.346,
"train_steps_per_second": 0.025
}