mistral_gsm8k_sft / all_results.json
weijie210's picture
Model save
b1bfa1b verified
{
"epoch": 1.0,
"eval_loss": 0.377094566822052,
"eval_runtime": 8.9964,
"eval_samples": 200,
"eval_samples_per_second": 22.231,
"eval_steps_per_second": 0.445,
"train_loss": 0.5052832438395574,
"train_runtime": 3169.7819,
"train_samples": 7273,
"train_samples_per_second": 2.294,
"train_steps_per_second": 0.144
}