初始化项目,由ModelHub XC社区提供模型

Model: babylm-anon/rta
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-06-05 15:06:17 +08:00
commit d1323322ae
13 changed files with 82824 additions and 0 deletions

9
train_results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"epoch": 9.728622631848438,
"total_flos": 7.9413964701696e+16,
"train_loss": 2.775124670731394,
"train_runtime": 3271.669,
"train_samples": 31240,
"train_samples_per_second": 92.919,
"train_steps_per_second": 5.807
}