| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.008965138752331828, | |
| "eval_steps": 100000000, | |
| "global_step": 1100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.15012613848348e-06, | |
| "grad_norm": 3.4710192680358887, | |
| "learning_rate": 1.0000000000000001e-07, | |
| "loss": 11.2605, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002445037841545044, | |
| "grad_norm": 5.464046955108643, | |
| "learning_rate": 3e-06, | |
| "loss": 10.9977, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0004890075683090088, | |
| "grad_norm": 1.1376756429672241, | |
| "learning_rate": 6e-06, | |
| "loss": 9.4082, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0007335113524635132, | |
| "grad_norm": 1.2878344058990479, | |
| "learning_rate": 9e-06, | |
| "loss": 8.549, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0009780151366180175, | |
| "grad_norm": 1.2800588607788086, | |
| "learning_rate": 1.2e-05, | |
| "loss": 8.0361, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.001222518920772522, | |
| "grad_norm": 1.4315314292907715, | |
| "learning_rate": 1.5e-05, | |
| "loss": 7.653, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0014670227049270264, | |
| "grad_norm": 1.381317377090454, | |
| "learning_rate": 1.8e-05, | |
| "loss": 7.4179, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.0017115264890815308, | |
| "grad_norm": 1.7989460229873657, | |
| "learning_rate": 2.1e-05, | |
| "loss": 7.2012, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.001956030273236035, | |
| "grad_norm": 1.3431414365768433, | |
| "learning_rate": 2.4e-05, | |
| "loss": 7.0383, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.0022005340573905395, | |
| "grad_norm": 1.028826117515564, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 6.866, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.002445037841545044, | |
| "grad_norm": 1.201025128364563, | |
| "learning_rate": 3e-05, | |
| "loss": 6.717, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.0026895416256995483, | |
| "grad_norm": 1.1023098230361938, | |
| "learning_rate": 3.3e-05, | |
| "loss": 6.5535, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.0029340454098540527, | |
| "grad_norm": 1.2839674949645996, | |
| "learning_rate": 3.6e-05, | |
| "loss": 6.4022, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.003178549194008557, | |
| "grad_norm": 2.267265796661377, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 6.2858, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.0034230529781630616, | |
| "grad_norm": 1.0635628700256348, | |
| "learning_rate": 4.2e-05, | |
| "loss": 6.1681, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.0036675567623175656, | |
| "grad_norm": 1.263838768005371, | |
| "learning_rate": 4.5e-05, | |
| "loss": 6.0728, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.00391206054647207, | |
| "grad_norm": 1.4611454010009766, | |
| "learning_rate": 4.8e-05, | |
| "loss": 5.972, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.0041565643306265745, | |
| "grad_norm": 1.0120561122894287, | |
| "learning_rate": 4.999999990869806e-05, | |
| "loss": 5.8619, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.004401068114781079, | |
| "grad_norm": 1.1349974870681763, | |
| "learning_rate": 4.999999853916893e-05, | |
| "loss": 5.785, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.004645571898935583, | |
| "grad_norm": 1.0840613842010498, | |
| "learning_rate": 4.9999995526204936e-05, | |
| "loss": 5.7071, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.004890075683090088, | |
| "grad_norm": 1.258074402809143, | |
| "learning_rate": 4.999999086980628e-05, | |
| "loss": 5.6199, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.005134579467244592, | |
| "grad_norm": 1.284726858139038, | |
| "learning_rate": 4.999998456997326e-05, | |
| "loss": 5.5465, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.005379083251399097, | |
| "grad_norm": 1.2079874277114868, | |
| "learning_rate": 4.999997662670628e-05, | |
| "loss": 5.4816, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.005623587035553601, | |
| "grad_norm": 1.3364052772521973, | |
| "learning_rate": 4.999996704000589e-05, | |
| "loss": 5.4079, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.0058680908197081055, | |
| "grad_norm": 0.9860705137252808, | |
| "learning_rate": 4.99999558098727e-05, | |
| "loss": 5.3598, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.00611259460386261, | |
| "grad_norm": 1.2071930170059204, | |
| "learning_rate": 4.9999942936307445e-05, | |
| "loss": 5.2884, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.006357098388017114, | |
| "grad_norm": 0.8959563970565796, | |
| "learning_rate": 4.9999928419310994e-05, | |
| "loss": 5.2391, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.006601602172171619, | |
| "grad_norm": 1.2356096506118774, | |
| "learning_rate": 4.999991225888427e-05, | |
| "loss": 5.1879, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.006846105956326123, | |
| "grad_norm": 0.9705113172531128, | |
| "learning_rate": 4.999989445502837e-05, | |
| "loss": 5.1424, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.007090609740480627, | |
| "grad_norm": 0.9504437446594238, | |
| "learning_rate": 4.9999875007744436e-05, | |
| "loss": 5.0966, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.007335113524635131, | |
| "grad_norm": 0.9488673806190491, | |
| "learning_rate": 4.9999853917033756e-05, | |
| "loss": 5.0424, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.007579617308789636, | |
| "grad_norm": 0.959373950958252, | |
| "learning_rate": 4.999983118289773e-05, | |
| "loss": 5.0387, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.00782412109294414, | |
| "grad_norm": 0.8465414643287659, | |
| "learning_rate": 4.999980680533782e-05, | |
| "loss": 4.9769, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.008068624877098645, | |
| "grad_norm": 0.8328993916511536, | |
| "learning_rate": 4.999978078435567e-05, | |
| "loss": 4.9335, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.008313128661253149, | |
| "grad_norm": 0.8107655644416809, | |
| "learning_rate": 4.999975311995295e-05, | |
| "loss": 4.9214, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.008557632445407654, | |
| "grad_norm": 0.8149654865264893, | |
| "learning_rate": 4.99997238121315e-05, | |
| "loss": 4.8651, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.008802136229562158, | |
| "grad_norm": 0.8837414979934692, | |
| "learning_rate": 4.999969286089325e-05, | |
| "loss": 4.8327, | |
| "step": 1080 | |
| } | |
| ], | |
| "logging_steps": 30, | |
| "max_steps": 368091, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1.052232627781632e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |