Model: xd2010/Qwen1.5-MOE-aux-free-sft-math7k-1e-3-gamma-1epoch Source: Original Platform
8 lines
217 B
JSON
8 lines
217 B
JSON
{
|
|
"total_flos": 2.0144468407196058e+17,
|
|
"train_loss": 0.3228219868138779,
|
|
"train_runtime": 1246.4994,
|
|
"train_samples": 6851,
|
|
"train_samples_per_second": 5.496,
|
|
"train_steps_per_second": 0.172
|
|
} |