Se124M100KInfPrompt_WT / train_results.json
augustocsc's picture
Model save
2d3a712 verified
{
"epoch": 3.0,
"total_flos": 1.4043296291217408e+16,
"train_loss": 0.9056532961688859,
"train_runtime": 6572.1588,
"train_samples_per_second": 35.683,
"train_steps_per_second": 1.115
}