SLM_vs_LLM_experiments
/
max_seq_length_128_experiments
/google_t5
/t5_small_patent
/train_results.json
{ | |
"epoch": 3.0, | |
"train_loss": 0.9882089091688776, | |
"train_runtime": 435.725, | |
"train_samples": 25000, | |
"train_samples_per_second": 172.127, | |
"train_steps_per_second": 5.384 | |
} |