jackf857/llama-3-8b-base-ipo-ultrafeedback-4xh200-batch-128-rerun-2-runpod
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: jackf857/llama-3-8b-base-ipo-ultrafeedback-4xh200-batch-128-rerun-2-runpod
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-05-26 12:46:17 +08:00
commit 8473c6f3cd
19 changed files with 3401 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
eval_results.json Normal file
View File

@@ -0,0 +1,16 @@
{
"epoch": 0.9989528795811519,
"eval_logits/chosen": -0.40516504645347595,
"eval_logits/rejected": -0.40924227237701416,
"eval_logps/chosen": -4.689553737640381,
"eval_logps/rejected": -7.291287422180176,
"eval_loss": 2328.408447265625,
"eval_rewards/accuracies": 0.6840000152587891,
"eval_rewards/chosen": -0.03572991117835045,
"eval_rewards/margins": 0.024208687245845795,
"eval_rewards/rejected": -0.05993859842419624,
"eval_runtime": 78.645,
"eval_samples": 2000,
"eval_samples_per_second": 25.431,
"eval_steps_per_second": 1.589
}