Files
CriticLeanGPT-Qwen2.5-7B-In…/train_results.json

10 lines
276 B
JSON
Raw Normal View History

{
"epoch": 3.0,
"tflops": 0.02876485648795673,
"token/s": 1971.4008815538489,
"total_flos": 372912231022592.0,
"train_loss": 0.24247261672442744,
"train_runtime": 12196.1154,
"train_samples_per_second": 11.689,
"train_steps_per_second": 0.731
}