Files
slm-125m-base/training_summary.json

9 lines
179 B
JSON
Raw Normal View History

{
"step": 4769,
"tokens_seen": 2500329472,
"target_tokens": 2500000000,
"max_steps": 0,
"world_size": 8,
"grad_accum_steps": 2,
"last_eval_loss": 2.303484795203459
}