Neelectric/Llama-3.1-8B-Instruct_SFT_mathsp_ewc_v00.07
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/Llama-3.1-8B-Instruct_SFT_mathsp_ewc_v00.07
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-06-04 18:30:45 +08:00
commit d6ce625128
17 changed files with 285717 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": 0.08365151286125183,
"ewc_loss_diag": 3.790855407714844e-05,
"ewc_loss_parallel": 0.00045809714356437325,
"total_flos": 5.62815163329864e+19,
"train_loss": 0.4756091437999748,
"train_runtime": 47393.4997,
"train_samples": 125770,
"train_samples_per_second": 7.961,
"train_steps_per_second": 0.498
}