| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.008150126138483479, | |
| "eval_steps": 100000000, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.15012613848348e-06, | |
| "grad_norm": 43.29102325439453, | |
| "learning_rate": 1.0000000000000001e-07, | |
| "loss": 81.0533, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002445037841545044, | |
| "grad_norm": 42.03493881225586, | |
| "learning_rate": 3e-06, | |
| "loss": 70.3325, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0004890075683090088, | |
| "grad_norm": 10.497812271118164, | |
| "learning_rate": 6e-06, | |
| "loss": 23.8844, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0007335113524635132, | |
| "grad_norm": 6.692718029022217, | |
| "learning_rate": 9e-06, | |
| "loss": 11.7198, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0009780151366180175, | |
| "grad_norm": 9.11446475982666, | |
| "learning_rate": 1.2e-05, | |
| "loss": 10.3199, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.001222518920772522, | |
| "grad_norm": 16.92243766784668, | |
| "learning_rate": 1.5e-05, | |
| "loss": 9.5338, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0014670227049270264, | |
| "grad_norm": 12.929216384887695, | |
| "learning_rate": 1.8e-05, | |
| "loss": 8.9114, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.0017115264890815308, | |
| "grad_norm": 8.830116271972656, | |
| "learning_rate": 2.1e-05, | |
| "loss": 8.3867, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.001956030273236035, | |
| "grad_norm": 7.348124980926514, | |
| "learning_rate": 2.4e-05, | |
| "loss": 8.0113, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.0022005340573905395, | |
| "grad_norm": 12.751787185668945, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 7.6093, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.002445037841545044, | |
| "grad_norm": 15.10444164276123, | |
| "learning_rate": 3e-05, | |
| "loss": 7.3484, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.0026895416256995483, | |
| "grad_norm": 6.653383731842041, | |
| "learning_rate": 3.3e-05, | |
| "loss": 7.2105, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.0029340454098540527, | |
| "grad_norm": 6.986039161682129, | |
| "learning_rate": 3.6e-05, | |
| "loss": 7.0228, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.003178549194008557, | |
| "grad_norm": 6.230088710784912, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 6.8948, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.0034230529781630616, | |
| "grad_norm": 8.170981407165527, | |
| "learning_rate": 4.2e-05, | |
| "loss": 6.6965, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.0036675567623175656, | |
| "grad_norm": 5.6345930099487305, | |
| "learning_rate": 4.5e-05, | |
| "loss": 6.5988, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.00391206054647207, | |
| "grad_norm": 5.156513214111328, | |
| "learning_rate": 4.8e-05, | |
| "loss": 6.4795, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.0041565643306265745, | |
| "grad_norm": 5.4964189529418945, | |
| "learning_rate": 4.999999990869806e-05, | |
| "loss": 6.333, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.004401068114781079, | |
| "grad_norm": 3.67378306388855, | |
| "learning_rate": 4.999999853916893e-05, | |
| "loss": 6.2208, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.004645571898935583, | |
| "grad_norm": 8.507222175598145, | |
| "learning_rate": 4.9999995526204936e-05, | |
| "loss": 6.1097, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.004890075683090088, | |
| "grad_norm": 3.756618022918701, | |
| "learning_rate": 4.999999086980628e-05, | |
| "loss": 5.9886, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.005134579467244592, | |
| "grad_norm": 3.8149781227111816, | |
| "learning_rate": 4.999998456997326e-05, | |
| "loss": 5.8779, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.005379083251399097, | |
| "grad_norm": 3.840543270111084, | |
| "learning_rate": 4.999997662670628e-05, | |
| "loss": 5.805, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.005623587035553601, | |
| "grad_norm": 3.4208931922912598, | |
| "learning_rate": 4.999996704000589e-05, | |
| "loss": 5.6992, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.0058680908197081055, | |
| "grad_norm": 3.2975683212280273, | |
| "learning_rate": 4.99999558098727e-05, | |
| "loss": 5.6531, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.00611259460386261, | |
| "grad_norm": 4.05631160736084, | |
| "learning_rate": 4.9999942936307445e-05, | |
| "loss": 5.554, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.006357098388017114, | |
| "grad_norm": 3.1539864540100098, | |
| "learning_rate": 4.9999928419310994e-05, | |
| "loss": 5.4931, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.006601602172171619, | |
| "grad_norm": 4.811732292175293, | |
| "learning_rate": 4.999991225888427e-05, | |
| "loss": 5.4204, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.006846105956326123, | |
| "grad_norm": 2.9593210220336914, | |
| "learning_rate": 4.999989445502837e-05, | |
| "loss": 5.3687, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.007090609740480627, | |
| "grad_norm": 3.942239284515381, | |
| "learning_rate": 4.9999875007744436e-05, | |
| "loss": 5.3238, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.007335113524635131, | |
| "grad_norm": 2.29752254486084, | |
| "learning_rate": 4.9999853917033756e-05, | |
| "loss": 5.2423, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.007579617308789636, | |
| "grad_norm": 2.243770122528076, | |
| "learning_rate": 4.999983118289773e-05, | |
| "loss": 5.2384, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.00782412109294414, | |
| "grad_norm": 2.5572686195373535, | |
| "learning_rate": 4.999980680533782e-05, | |
| "loss": 5.1761, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.008068624877098645, | |
| "grad_norm": 2.4739913940429688, | |
| "learning_rate": 4.999978078435567e-05, | |
| "loss": 5.1215, | |
| "step": 990 | |
| } | |
| ], | |
| "logging_steps": 30, | |
| "max_steps": 368091, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 9.730730754048e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |