SLM_vs_LLM_experiments
/
max_seq_length_128_experiments
/microsoft
/phi_2_patent
/train_results.json
{ | |
"epoch": 3.0, | |
"train_loss": 1.123963495044757, | |
"train_runtime": 3214.4035, | |
"train_samples": 25000, | |
"train_samples_per_second": 23.332, | |
"train_steps_per_second": 0.365 | |
} |