Files
OLMo-2-1124-7B-Instruct_SFT…/train_results.json

11 lines
343 B
JSON
Raw Normal View History

{
"ewc_loss": 0.0810546875,
"ewc_loss_diag": 0.0003795623779296875,
"ewc_loss_parallel": 4.3392181396484375e-05,
"total_flos": 5.156325133756924e+19,
"train_loss": 0.44972702040424817,
"train_runtime": 50365.0092,
"train_samples": 125770,
"train_samples_per_second": 7.492,
"train_steps_per_second": 0.468
}