{ "effective_tokens_per_sec": 4074.350224923108, "epoch": 1.0, "total_flos": 8.183535699912294e+17, "train_loss": 1.4031060467437138, "train_runtime": 3980.916, "train_samples_per_second": 2.629, "train_steps_per_second": 0.082 }