SLM_vs_LLM_experiments
/
max_seq_length_128_experiments
/distilbert
/distilroberta_base_amazon
/all_results.json
{ | |
"epoch": 3.0, | |
"eval_accuracy": 0.782608695652174, | |
"eval_f1_macro": 0.713683768180215, | |
"eval_f1_micro": 0.782608695652174, | |
"eval_loss": 0.777472734451294, | |
"eval_runtime": 0.8146, | |
"eval_samples": 1518, | |
"eval_samples_per_second": 1863.389, | |
"eval_steps_per_second": 29.461, | |
"test_accuracy": 0.7990777338603425, | |
"test_f1_macro": 0.7328875867864995, | |
"test_f1_micro": 0.7990777338603425, | |
"test_loss": 0.7488571405410767, | |
"test_runtime": 0.8133, | |
"test_samples_per_second": 1866.563, | |
"test_steps_per_second": 29.511, | |
"train_loss": 1.1344423653786642, | |
"train_runtime": 82.3443, | |
"train_samples": 12144, | |
"train_samples_per_second": 442.435, | |
"train_steps_per_second": 6.922 | |
} |