SLM_vs_LLM_experiments
/
max_seq_length_128_experiments
/distilbert
/distilroberta_base_ledgar
/train_results.json
{ | |
"epoch": 3.0, | |
"train_loss": 1.1111405622730377, | |
"train_runtime": 461.0061, | |
"train_samples": 60000, | |
"train_samples_per_second": 390.45, | |
"train_steps_per_second": 6.104 | |
} |