Files
CriticLeanGPT-Qwen2.5-14B-I…/all_results.json

14 lines
416 B
JSON
Raw Permalink Normal View History

{
"epoch": 3.0,
"eval_loss": 0.29699257016181946,
"eval_runtime": 15.9659,
"eval_samples_per_second": 30.064,
"eval_steps_per_second": 1.879,
"tflops": 0.07104939945192446,
"token/s": 964.1313936883596,
"total_flos": 940586567729152.0,
"train_loss": 0.23212839591115145,
"train_runtime": 12724.6535,
"train_samples_per_second": 11.203,
"train_steps_per_second": 0.35
}