初始化项目,由ModelHub XC社区提供模型

Model: open-sci/sft__ot30k_Qwen2.5-1.5B-SFT-Tulu3-decontaminated
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-04-26 22:30:08 +08:00
commit 5326c0f042
20 changed files with 166071 additions and 0 deletions

12
train_results.json Normal file
View File

@@ -0,0 +1,12 @@
{
"epoch": 5.0,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.4849323630332947,
"total_flos": 1940656491724800.0,
"train_loss": 1.0709928606926127,
"train_runtime": 5219.2407,
"train_samples_per_second": 28.74,
"train_steps_per_second": 0.225,
"valid_targets_mean": 11353.3,
"valid_targets_min": 2084
}