SLM_vs_LLM_experiments
/
max_seq_length_128_experiments
/distilbert
/distilroberta_base_patent
/train_results.json
{ | |
"epoch": 3.0, | |
"train_loss": 1.0698413495212564, | |
"train_runtime": 210.2936, | |
"train_samples": 25000, | |
"train_samples_per_second": 356.644, | |
"train_steps_per_second": 5.578 | |
} |