| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.666666666666667, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 6.473907810784441, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8182, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 6.023865943741977, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.7717, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 5.721027974314376, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.8082, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 2.368283355512373, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7099, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 3.785435757041654, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7267, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 4.547368154663255, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7358, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 4.518316807995772, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6916, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 2.6583778204111743, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.666, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.402750179268395, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6408, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 1.9453807161447727, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.6254, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 2.1403080295084522, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9023, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.2160713982841074, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5659, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 1.0756802880114802, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.5446, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 1.1092122804496283, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.568, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.8045414037997163, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5098, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 0.8378971998896949, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5541, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.7874041965605373, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5321, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.8861776328889852, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.544, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.7666998483784012, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.5178, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.9245134158891382, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.5232, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.8716106177309219, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.8009, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.7646836628120641, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4936, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.635291872606153, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4832, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.6111150929455261, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4644, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.6345576605484466, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4673, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.5357012677333587, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.4724, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.5833576917610517, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4693, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.5392935393590009, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.465, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.49903504138085425, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4606, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.5305436427031079, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.4638, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.9523809523809526, | |
| "grad_norm": 0.528894706931349, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.4597, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.0476190476190474, | |
| "grad_norm": 0.7013576018441137, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6691, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.48458458030385704, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4262, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.238095238095238, | |
| "grad_norm": 0.5156224435213471, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4336, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.4192384327361703, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4202, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.4679317485041411, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.4112, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.5238095238095237, | |
| "grad_norm": 0.423350161647943, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.403, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.619047619047619, | |
| "grad_norm": 0.4253952845422581, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.424, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.4406922753405612, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4228, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.3650694481105728, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3864, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.9047619047619047, | |
| "grad_norm": 0.3389421147434116, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.4163, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5423871607040233, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.63, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.095238095238095, | |
| "grad_norm": 0.33711740698958864, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3931, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 0.3707565651114291, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3853, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.3297542280002762, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3858, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 0.32728774546919065, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3918, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.476190476190476, | |
| "grad_norm": 0.31370608092833757, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.378, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.2954983993539471, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.388, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.3186767235875737, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3709, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.2633111193930405, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3798, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 0.2733416484025912, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3566, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.9523809523809526, | |
| "grad_norm": 0.29719699722412773, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.3854, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.0476190476190474, | |
| "grad_norm": 0.3833070305252553, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.544, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.2421660534659939, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3788, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.238095238095238, | |
| "grad_norm": 0.23162689078219867, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.352, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.24819293407328777, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3642, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 0.24490566955492518, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.3546, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.523809523809524, | |
| "grad_norm": 0.2565537750302216, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.355, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.619047619047619, | |
| "grad_norm": 0.23803146329314778, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.3779, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.24360029567807934, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.3552, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.809523809523809, | |
| "grad_norm": 0.22145534531153122, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.3553, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.904761904761905, | |
| "grad_norm": 0.22076587627820649, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3682, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.3353631406460431, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.548, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.095238095238095, | |
| "grad_norm": 0.20615267336743034, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3655, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.190476190476191, | |
| "grad_norm": 0.20334262398668054, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.3583, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.21998318653368576, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.3602, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.380952380952381, | |
| "grad_norm": 0.2075736200574408, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.3503, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.476190476190476, | |
| "grad_norm": 0.2081878657542074, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.3555, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 0.20849929219392016, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.3613, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.1963311786223033, | |
| "learning_rate": 0.0, | |
| "loss": 0.341, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "step": 70, | |
| "total_flos": 2.4044071957823488e+17, | |
| "train_loss": 0.48798870486872536, | |
| "train_runtime": 2935.3937, | |
| "train_samples_per_second": 2.385, | |
| "train_steps_per_second": 0.024 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 70, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.4044071957823488e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |