| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 410, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24390243902439024, | |
| "grad_norm": 7.625, | |
| "learning_rate": 0.00019970658011837404, | |
| "loss": 2.0086, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4878048780487805, | |
| "grad_norm": 0.59033203125, | |
| "learning_rate": 0.00019882804237803488, | |
| "loss": 1.1443, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7317073170731707, | |
| "grad_norm": 0.736328125, | |
| "learning_rate": 0.00019736954238777792, | |
| "loss": 0.985, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.975609756097561, | |
| "grad_norm": 0.50048828125, | |
| "learning_rate": 0.00019533963920549306, | |
| "loss": 0.918, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "grad_norm": 0.36083984375, | |
| "learning_rate": 0.0001927502451102095, | |
| "loss": 0.7771, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.4634146341463414, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 0.00018961655569610557, | |
| "loss": 0.8079, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.7073170731707317, | |
| "grad_norm": 0.314208984375, | |
| "learning_rate": 0.00018595696069872013, | |
| "loss": 0.7491, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.951219512195122, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 0.00018179293607667178, | |
| "loss": 0.7349, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.1951219512195124, | |
| "grad_norm": 0.331787109375, | |
| "learning_rate": 0.0001771489179821943, | |
| "loss": 0.7468, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "grad_norm": 0.374755859375, | |
| "learning_rate": 0.0001720521593600787, | |
| "loss": 0.7582, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.682926829268293, | |
| "grad_norm": 0.3662109375, | |
| "learning_rate": 0.00016653257001655652, | |
| "loss": 0.715, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.926829268292683, | |
| "grad_norm": 0.413818359375, | |
| "learning_rate": 0.0001606225410966638, | |
| "loss": 0.7404, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.1707317073170733, | |
| "grad_norm": 0.349609375, | |
| "learning_rate": 0.00015435675500012212, | |
| "loss": 0.6844, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.4146341463414633, | |
| "grad_norm": 0.48095703125, | |
| "learning_rate": 0.0001477719818512263, | |
| "loss": 0.7568, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.658536585365854, | |
| "grad_norm": 0.369384765625, | |
| "learning_rate": 0.00014090686371713402, | |
| "loss": 0.6647, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.902439024390244, | |
| "grad_norm": 0.422607421875, | |
| "learning_rate": 0.00013380168784085027, | |
| "loss": 0.698, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.146341463414634, | |
| "grad_norm": 0.38427734375, | |
| "learning_rate": 0.0001264981502196662, | |
| "loss": 0.6861, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.390243902439025, | |
| "grad_norm": 0.38525390625, | |
| "learning_rate": 0.00011903911091646684, | |
| "loss": 0.6855, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.634146341463414, | |
| "grad_norm": 0.384033203125, | |
| "learning_rate": 0.00011146834253984006, | |
| "loss": 0.6797, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.878048780487805, | |
| "grad_norm": 0.365966796875, | |
| "learning_rate": 0.00010383027336900355, | |
| "loss": 0.6865, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.121951219512195, | |
| "grad_norm": 0.39794921875, | |
| "learning_rate": 9.616972663099647e-05, | |
| "loss": 0.6859, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.365853658536586, | |
| "grad_norm": 0.3837890625, | |
| "learning_rate": 8.853165746015997e-05, | |
| "loss": 0.6636, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.609756097560975, | |
| "grad_norm": 0.445068359375, | |
| "learning_rate": 8.096088908353315e-05, | |
| "loss": 0.6883, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.853658536585366, | |
| "grad_norm": 0.4130859375, | |
| "learning_rate": 7.350184978033386e-05, | |
| "loss": 0.6432, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.097560975609756, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 6.619831215914974e-05, | |
| "loss": 0.6597, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.341463414634147, | |
| "grad_norm": 0.407958984375, | |
| "learning_rate": 5.909313628286601e-05, | |
| "loss": 0.6574, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.585365853658536, | |
| "grad_norm": 0.3974609375, | |
| "learning_rate": 5.222801814877369e-05, | |
| "loss": 0.6499, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.829268292682927, | |
| "grad_norm": 0.40625, | |
| "learning_rate": 4.56432449998779e-05, | |
| "loss": 0.6436, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.073170731707317, | |
| "grad_norm": 0.3994140625, | |
| "learning_rate": 3.937745890333623e-05, | |
| "loss": 0.6644, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.317073170731708, | |
| "grad_norm": 0.40673828125, | |
| "learning_rate": 3.346742998344348e-05, | |
| "loss": 0.656, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.560975609756097, | |
| "grad_norm": 0.408447265625, | |
| "learning_rate": 2.794784063992131e-05, | |
| "loss": 0.6244, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.804878048780488, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 2.2851082017805703e-05, | |
| "loss": 0.6594, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.048780487804878, | |
| "grad_norm": 0.437744140625, | |
| "learning_rate": 1.8207063923328237e-05, | |
| "loss": 0.6423, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.292682926829269, | |
| "grad_norm": 0.43212890625, | |
| "learning_rate": 1.4043039301279903e-05, | |
| "loss": 0.635, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.536585365853659, | |
| "grad_norm": 0.4013671875, | |
| "learning_rate": 1.0383444303894452e-05, | |
| "loss": 0.6408, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.78048780487805, | |
| "grad_norm": 0.40869140625, | |
| "learning_rate": 7.249754889790539e-06, | |
| "loss": 0.6438, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.024390243902438, | |
| "grad_norm": 0.3857421875, | |
| "learning_rate": 4.660360794506946e-06, | |
| "loss": 0.6282, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.268292682926829, | |
| "grad_norm": 0.400634765625, | |
| "learning_rate": 2.6304576122221035e-06, | |
| "loss": 0.6359, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.512195121951219, | |
| "grad_norm": 0.3935546875, | |
| "learning_rate": 1.1719576219651585e-06, | |
| "loss": 0.6467, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.75609756097561, | |
| "grad_norm": 0.4140625, | |
| "learning_rate": 2.934198816259559e-07, | |
| "loss": 0.6479, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.400390625, | |
| "learning_rate": 0.0, | |
| "loss": 0.6203, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 410, | |
| "total_flos": 2.000692923334656e+16, | |
| "train_loss": 0.7381390141277778, | |
| "train_runtime": 366.1191, | |
| "train_samples_per_second": 4.479, | |
| "train_steps_per_second": 1.12 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 410, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2.000692923334656e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |