SLM_vs_LLM_experiments
/
max_seq_length_128_experiments
/google_t5
/t5_base_patent
/train_results.json
{ | |
"epoch": 3.0, | |
"train_loss": 0.8408535616273872, | |
"train_runtime": 1368.6484, | |
"train_samples": 25000, | |
"train_samples_per_second": 54.799, | |
"train_steps_per_second": 1.714 | |
} |