SLM_vs_LLM_experiments
/
max_seq_length_128_experiments
/distilbert
/distilbert_base_uncased_amazon
/all_results.json
{ | |
"epoch": 3.0, | |
"eval_accuracy": 0.7575757575757576, | |
"eval_f1_macro": 0.6903636946713366, | |
"eval_f1_micro": 0.7575757575757576, | |
"eval_loss": 0.9129917025566101, | |
"eval_runtime": 0.8205, | |
"eval_samples": 1518, | |
"eval_samples_per_second": 1850.15, | |
"eval_steps_per_second": 29.251, | |
"test_accuracy": 0.7977602108036891, | |
"test_f1_macro": 0.7193918798114766, | |
"test_f1_micro": 0.7977602108036891, | |
"test_loss": 0.8355989456176758, | |
"test_runtime": 0.8172, | |
"test_samples_per_second": 1857.675, | |
"test_steps_per_second": 29.37, | |
"train_loss": 1.337625675870661, | |
"train_runtime": 76.5319, | |
"train_samples": 12144, | |
"train_samples_per_second": 476.037, | |
"train_steps_per_second": 7.448 | |
} |