| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 43.146067415730336, | |
| "eval_steps": 500, | |
| "global_step": 960, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 7.640738010406494, | |
| "learning_rate": 0.0002, | |
| "loss": 4.2661, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 2.9313442707061768, | |
| "learning_rate": 0.0002, | |
| "loss": 2.306, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 0.9562789797782898, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0093, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 1.4072426557540894, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8391, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "grad_norm": 1.1522574424743652, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7152, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "grad_norm": 0.9669118523597717, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5656, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "grad_norm": 1.1254830360412598, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4272, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "grad_norm": 1.4298350811004639, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2586, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "grad_norm": 1.3048124313354492, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1003, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "grad_norm": 1.5784626007080078, | |
| "learning_rate": 0.0002, | |
| "loss": 0.941, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "grad_norm": 1.6184762716293335, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7854, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.79, | |
| "grad_norm": 2.039607048034668, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6663, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 11.69, | |
| "grad_norm": 1.9069631099700928, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5598, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.58, | |
| "grad_norm": 3.269792079925537, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4746, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 13.48, | |
| "grad_norm": 1.695237159729004, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3884, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.38, | |
| "grad_norm": 1.7961617708206177, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3197, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 15.28, | |
| "grad_norm": 1.6906554698944092, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2876, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.18, | |
| "grad_norm": 1.5338362455368042, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2476, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 17.08, | |
| "grad_norm": 1.482823371887207, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2152, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 17.98, | |
| "grad_norm": 1.6050206422805786, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1864, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 18.88, | |
| "grad_norm": 1.7870419025421143, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1527, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 19.78, | |
| "grad_norm": 1.6181118488311768, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1387, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 20.67, | |
| "grad_norm": 1.545577049255371, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1291, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 21.57, | |
| "grad_norm": 1.4766790866851807, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1216, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 22.47, | |
| "grad_norm": 1.2652430534362793, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1125, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 23.37, | |
| "grad_norm": 1.3792601823806763, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1064, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 24.27, | |
| "grad_norm": 1.1617250442504883, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0965, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 25.17, | |
| "grad_norm": 1.0318264961242676, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0892, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 26.07, | |
| "grad_norm": 1.0102779865264893, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0866, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "grad_norm": 1.2883203029632568, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0805, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 27.87, | |
| "grad_norm": 1.1580032110214233, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0748, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 28.76, | |
| "grad_norm": 1.114597201347351, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0745, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 29.66, | |
| "grad_norm": 1.0546940565109253, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0721, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 30.56, | |
| "grad_norm": 1.0050326585769653, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0697, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 31.46, | |
| "grad_norm": 1.0160025358200073, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0658, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 32.36, | |
| "grad_norm": 0.9212460517883301, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0643, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 33.26, | |
| "grad_norm": 0.8616517186164856, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0621, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 34.16, | |
| "grad_norm": 0.8040679693222046, | |
| "learning_rate": 0.0002, | |
| "loss": 0.061, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 35.06, | |
| "grad_norm": 0.7591003179550171, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0589, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 35.96, | |
| "grad_norm": 1.0100669860839844, | |
| "learning_rate": 0.0002, | |
| "loss": 0.055, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 36.85, | |
| "grad_norm": 0.9912341237068176, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0513, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 37.75, | |
| "grad_norm": 0.9290223121643066, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0515, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 38.65, | |
| "grad_norm": 0.8802034854888916, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0511, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 39.55, | |
| "grad_norm": 0.8020614981651306, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0506, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 40.45, | |
| "grad_norm": 0.8280277848243713, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0502, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 41.35, | |
| "grad_norm": 0.7979443669319153, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0496, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 42.25, | |
| "grad_norm": 0.7503458857536316, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0487, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 43.15, | |
| "grad_norm": 0.7012200355529785, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0469, | |
| "step": 960 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 46, | |
| "save_steps": 20, | |
| "total_flos": 2.927218722298921e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |