{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 63, "global_step": 63, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.047619047619047616, "grad_norm": 8.636988639831543, "learning_rate": 1e-05, "loss": 2.1923828125, "nan_loss_batches": 0, "step": 1 }, { "epoch": 0.09523809523809523, "grad_norm": 6.270175457000732, "learning_rate": 9.993784606094612e-06, "loss": 1.861328125, "nan_loss_batches": 0, "step": 2 }, { "epoch": 0.14285714285714285, "grad_norm": 12.795247077941895, "learning_rate": 9.975153876827008e-06, "loss": 1.7330322265625, "nan_loss_batches": 0, "step": 3 }, { "epoch": 0.19047619047619047, "grad_norm": 4.768237590789795, "learning_rate": 9.944154131125643e-06, "loss": 1.62890625, "nan_loss_batches": 0, "step": 4 }, { "epoch": 0.23809523809523808, "grad_norm": 3.3463194370269775, "learning_rate": 9.900862439242719e-06, "loss": 1.60986328125, "nan_loss_batches": 0, "step": 5 }, { "epoch": 0.2857142857142857, "grad_norm": 1.99888014793396, "learning_rate": 9.84538643114539e-06, "loss": 1.537109375, "nan_loss_batches": 0, "step": 6 }, { "epoch": 0.3333333333333333, "grad_norm": 1.9878199100494385, "learning_rate": 9.777864028930705e-06, "loss": 1.49609375, "nan_loss_batches": 0, "step": 7 }, { "epoch": 0.38095238095238093, "grad_norm": 1.9544587135314941, "learning_rate": 9.698463103929542e-06, "loss": 1.46728515625, "nan_loss_batches": 0, "step": 8 }, { "epoch": 0.42857142857142855, "grad_norm": 1.8168585300445557, "learning_rate": 9.60738105935204e-06, "loss": 1.4150390625, "nan_loss_batches": 0, "step": 9 }, { "epoch": 0.47619047619047616, "grad_norm": 1.6426150798797607, "learning_rate": 9.504844339512096e-06, "loss": 1.42041015625, "nan_loss_batches": 0, "step": 10 }, { "epoch": 0.5238095238095238, "grad_norm": 1.5644619464874268, "learning_rate": 9.391107866851143e-06, "loss": 1.4189453125, "nan_loss_batches": 0, "step": 11 }, { "epoch": 0.5714285714285714, "grad_norm": 1.5254902839660645, "learning_rate": 9.266454408160779e-06, "loss": 1.4052734375, "nan_loss_batches": 0, "step": 12 }, { "epoch": 0.6190476190476191, "grad_norm": 1.4199761152267456, "learning_rate": 9.131193871579975e-06, "loss": 1.38427734375, "nan_loss_batches": 0, "step": 13 }, { "epoch": 0.6666666666666666, "grad_norm": 1.4111160039901733, "learning_rate": 8.985662536114614e-06, "loss": 1.41748046875, "nan_loss_batches": 0, "step": 14 }, { "epoch": 0.7142857142857143, "grad_norm": 1.5193932056427002, "learning_rate": 8.83022221559489e-06, "loss": 1.35546875, "nan_loss_batches": 0, "step": 15 }, { "epoch": 0.7619047619047619, "grad_norm": 1.3467028141021729, "learning_rate": 8.665259359149132e-06, "loss": 1.35205078125, "nan_loss_batches": 0, "step": 16 }, { "epoch": 0.8095238095238095, "grad_norm": 1.3716109991073608, "learning_rate": 8.491184090430365e-06, "loss": 1.357421875, "nan_loss_batches": 0, "step": 17 }, { "epoch": 0.8571428571428571, "grad_norm": 1.3842028379440308, "learning_rate": 8.308429187984298e-06, "loss": 1.365234375, "nan_loss_batches": 0, "step": 18 }, { "epoch": 0.9047619047619048, "grad_norm": 1.1998069286346436, "learning_rate": 8.117449009293668e-06, "loss": 1.314453125, "nan_loss_batches": 0, "step": 19 }, { "epoch": 0.9523809523809523, "grad_norm": 1.4626988172531128, "learning_rate": 7.918718361173951e-06, "loss": 1.34814453125, "nan_loss_batches": 0, "step": 20 }, { "epoch": 1.0, "grad_norm": 1.3381778001785278, "learning_rate": 7.712731319328798e-06, "loss": 1.30224609375, "nan_loss_batches": 0, "step": 21 }, { "epoch": 1.0476190476190477, "grad_norm": 1.3117152452468872, "learning_rate": 7.500000000000001e-06, "loss": 1.27001953125, "nan_loss_batches": 0, "step": 22 }, { "epoch": 1.0952380952380953, "grad_norm": 1.3333078622817993, "learning_rate": 7.281053286765816e-06, "loss": 1.2587890625, "nan_loss_batches": 0, "step": 23 }, { "epoch": 1.1428571428571428, "grad_norm": 1.4496302604675293, "learning_rate": 7.056435515653059e-06, "loss": 1.27593994140625, "nan_loss_batches": 0, "step": 24 }, { "epoch": 1.1904761904761905, "grad_norm": 1.397755742073059, "learning_rate": 6.8267051218319766e-06, "loss": 1.2216796875, "nan_loss_batches": 0, "step": 25 }, { "epoch": 1.2380952380952381, "grad_norm": 1.281689167022705, "learning_rate": 6.592433251258423e-06, "loss": 1.28662109375, "nan_loss_batches": 0, "step": 26 }, { "epoch": 1.2857142857142856, "grad_norm": 1.411041259765625, "learning_rate": 6.354202340715027e-06, "loss": 1.146484375, "nan_loss_batches": 0, "step": 27 }, { "epoch": 1.3333333333333333, "grad_norm": 1.367311716079712, "learning_rate": 6.112604669781572e-06, "loss": 1.127197265625, "nan_loss_batches": 0, "step": 28 }, { "epoch": 1.380952380952381, "grad_norm": 1.2573864459991455, "learning_rate": 5.8682408883346535e-06, "loss": 1.17822265625, "nan_loss_batches": 0, "step": 29 }, { "epoch": 1.4285714285714286, "grad_norm": 1.459160566329956, "learning_rate": 5.621718523237427e-06, "loss": 1.0859375, "nan_loss_batches": 0, "step": 30 }, { "epoch": 1.4761904761904763, "grad_norm": 1.3222463130950928, "learning_rate": 5.373650467932122e-06, "loss": 1.12060546875, "nan_loss_batches": 0, "step": 31 }, { "epoch": 1.5238095238095237, "grad_norm": 1.264496088027954, "learning_rate": 5.1246534586903655e-06, "loss": 1.1337890625, "nan_loss_batches": 0, "step": 32 }, { "epoch": 1.5714285714285714, "grad_norm": 1.394464373588562, "learning_rate": 4.875346541309637e-06, "loss": 1.119873046875, "nan_loss_batches": 0, "step": 33 }, { "epoch": 1.619047619047619, "grad_norm": 1.3051438331604004, "learning_rate": 4.626349532067879e-06, "loss": 1.09228515625, "nan_loss_batches": 0, "step": 34 }, { "epoch": 1.6666666666666665, "grad_norm": 1.2042930126190186, "learning_rate": 4.3782814767625755e-06, "loss": 1.16943359375, "nan_loss_batches": 0, "step": 35 }, { "epoch": 1.7142857142857144, "grad_norm": 1.35514235496521, "learning_rate": 4.131759111665349e-06, "loss": 1.08642578125, "nan_loss_batches": 0, "step": 36 }, { "epoch": 1.7619047619047619, "grad_norm": 1.262746810913086, "learning_rate": 3.887395330218429e-06, "loss": 1.123046875, "nan_loss_batches": 0, "step": 37 }, { "epoch": 1.8095238095238095, "grad_norm": 1.2942466735839844, "learning_rate": 3.6457976592849753e-06, "loss": 1.1103515625, "nan_loss_batches": 0, "step": 38 }, { "epoch": 1.8571428571428572, "grad_norm": 1.2770276069641113, "learning_rate": 3.4075667487415785e-06, "loss": 1.14501953125, "nan_loss_batches": 0, "step": 39 }, { "epoch": 1.9047619047619047, "grad_norm": 1.1975665092468262, "learning_rate": 3.173294878168025e-06, "loss": 1.09228515625, "nan_loss_batches": 0, "step": 40 }, { "epoch": 1.9523809523809523, "grad_norm": 1.277531385421753, "learning_rate": 2.9435644843469434e-06, "loss": 1.138671875, "nan_loss_batches": 0, "step": 41 }, { "epoch": 2.0, "grad_norm": 1.4103481769561768, "learning_rate": 2.718946713234185e-06, "loss": 1.01318359375, "nan_loss_batches": 0, "step": 42 }, { "epoch": 2.0476190476190474, "grad_norm": 1.271584391593933, "learning_rate": 2.5000000000000015e-06, "loss": 1.01611328125, "nan_loss_batches": 0, "step": 43 }, { "epoch": 2.0952380952380953, "grad_norm": 1.2585103511810303, "learning_rate": 2.2872686806712037e-06, "loss": 1.083984375, "nan_loss_batches": 0, "step": 44 }, { "epoch": 2.142857142857143, "grad_norm": 1.4265955686569214, "learning_rate": 2.081281638826052e-06, "loss": 1.0775146484375, "nan_loss_batches": 0, "step": 45 }, { "epoch": 2.1904761904761907, "grad_norm": 1.3593528270721436, "learning_rate": 1.8825509907063328e-06, "loss": 1.0439453125, "nan_loss_batches": 0, "step": 46 }, { "epoch": 2.238095238095238, "grad_norm": 1.2724002599716187, "learning_rate": 1.6915708120157042e-06, "loss": 1.14404296875, "nan_loss_batches": 0, "step": 47 }, { "epoch": 2.2857142857142856, "grad_norm": 1.3185619115829468, "learning_rate": 1.5088159095696365e-06, "loss": 0.99365234375, "nan_loss_batches": 0, "step": 48 }, { "epoch": 2.3333333333333335, "grad_norm": 1.3230262994766235, "learning_rate": 1.3347406408508695e-06, "loss": 0.973876953125, "nan_loss_batches": 0, "step": 49 }, { "epoch": 2.380952380952381, "grad_norm": 1.1950162649154663, "learning_rate": 1.1697777844051105e-06, "loss": 1.05224609375, "nan_loss_batches": 0, "step": 50 }, { "epoch": 2.4285714285714284, "grad_norm": 1.3244589567184448, "learning_rate": 1.0143374638853892e-06, "loss": 0.9501953125, "nan_loss_batches": 0, "step": 51 }, { "epoch": 2.4761904761904763, "grad_norm": 1.2009601593017578, "learning_rate": 8.688061284200266e-07, "loss": 0.99169921875, "nan_loss_batches": 0, "step": 52 }, { "epoch": 2.5238095238095237, "grad_norm": 1.1781463623046875, "learning_rate": 7.33545591839222e-07, "loss": 1.009765625, "nan_loss_batches": 0, "step": 53 }, { "epoch": 2.571428571428571, "grad_norm": 1.223464012145996, "learning_rate": 6.088921331488568e-07, "loss": 1.00244140625, "nan_loss_batches": 0, "step": 54 }, { "epoch": 2.619047619047619, "grad_norm": 1.2199597358703613, "learning_rate": 4.951556604879049e-07, "loss": 0.97607421875, "nan_loss_batches": 0, "step": 55 }, { "epoch": 2.6666666666666665, "grad_norm": 1.1372495889663696, "learning_rate": 3.9261894064796136e-07, "loss": 1.0673828125, "nan_loss_batches": 0, "step": 56 }, { "epoch": 2.7142857142857144, "grad_norm": 1.3241941928863525, "learning_rate": 3.015368960704584e-07, "loss": 0.9814453125, "nan_loss_batches": 0, "step": 57 }, { "epoch": 2.761904761904762, "grad_norm": 1.1642942428588867, "learning_rate": 2.2213597106929608e-07, "loss": 1.03955078125, "nan_loss_batches": 0, "step": 58 }, { "epoch": 2.8095238095238093, "grad_norm": 1.1905473470687866, "learning_rate": 1.5461356885461077e-07, "loss": 1.02001953125, "nan_loss_batches": 0, "step": 59 }, { "epoch": 2.857142857142857, "grad_norm": 1.16309654712677, "learning_rate": 9.913756075728088e-08, "loss": 1.0634765625, "nan_loss_batches": 0, "step": 60 }, { "epoch": 2.9047619047619047, "grad_norm": 1.1112571954727173, "learning_rate": 5.584586887435739e-08, "loss": 1.0234375, "nan_loss_batches": 0, "step": 61 }, { "epoch": 2.9523809523809526, "grad_norm": 1.2206873893737793, "learning_rate": 2.4846123172992953e-08, "loss": 1.06591796875, "nan_loss_batches": 0, "step": 62 }, { "epoch": 3.0, "grad_norm": 1.2536906003952026, "learning_rate": 6.215393905388278e-09, "loss": 0.930419921875, "nan_loss_batches": 0, "step": 63 }, { "epoch": 3.0, "eval_loss": 1.5146484375, "eval_nan_loss_batches": 0, "eval_runtime": 12.9822, "eval_samples_per_second": 0.231, "eval_steps_per_second": 0.077, "step": 63 } ], "logging_steps": 1.0, "max_steps": 63, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 0, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 2.3323949743158067e+17, "train_batch_size": 1, "trial_name": null, "trial_params": null }