Neelectric/Ministral-8B-Instruct-2410_SFT_mathsp_ewc_v00.01
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: Neelectric/Ministral-8B-Instruct-2410_SFT_mathsp_ewc_v00.01
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-09-27 14:32:07 +08:00
commit 141a8c8c5e
17 changed files with 291712 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
train_results.json Normal file
View File

@@ -0,0 +1,11 @@
{
"ewc_loss": 0.023193359375,
"ewc_loss_diag": 0.0001049041748046875,
"ewc_loss_parallel": 1.2636184692382812e-05,
"total_flos": 5.751448627479105e+19,
"train_loss": 0.3408017477382914,
"train_runtime": 50265.1743,
"train_samples": 125770,
"train_samples_per_second": 7.506,
"train_steps_per_second": 0.469
}