| { | |
| "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": 6.605348131730995, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8419, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 6.421236878123063, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8347, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 5.6301912031095, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.7892, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 2.3481103388632967, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7191, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 3.77462144106806, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7275, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 4.145002917098403, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7245, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 4.160357161281431, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6869, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 2.530895080447052, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6701, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.2651868042761207, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6473, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 2.017603532790786, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.6124, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 2.142230908069242, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9323, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.192411410964399, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5877, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 1.1055427383462693, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.5601, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 1.0654694223469574, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5828, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.9654479013607004, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5557, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 0.8914056609525524, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5344, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.9774447718066034, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5558, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.7790629886541486, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.5179, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.8202964518997443, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.529, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.8155377516615383, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.528, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.0671953476376677, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.7894, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.5021207833915019, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4923, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.6991883919562059, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4884, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.5720037808635232, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4665, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.5458417578805382, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4782, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.7366558111251066, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.4858, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.5385737455534967, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4796, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.47977596305770687, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.4516, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.5155484599482115, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4749, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.44464289664292356, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.467, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.9523809523809526, | |
| "grad_norm": 0.5523193408133361, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.4706, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.0476190476190474, | |
| "grad_norm": 0.5264144060681053, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6569, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.41126779074862735, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4292, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.238095238095238, | |
| "grad_norm": 0.4490298708496658, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4378, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.3886114013945829, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4262, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.38569867397130814, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.4253, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.5238095238095237, | |
| "grad_norm": 0.32667953259390436, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4342, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.619047619047619, | |
| "grad_norm": 0.34387926938768176, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4293, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.37550697839247893, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.416, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.3339790028645127, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.423, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.9047619047619047, | |
| "grad_norm": 0.30613589011704845, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.4052, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5533136576839677, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.6213, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.095238095238095, | |
| "grad_norm": 0.31266304108004894, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3984, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 0.333588488447849, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3867, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.3460065169199803, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3847, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 0.30272468877509284, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.381, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.476190476190476, | |
| "grad_norm": 0.3046448293769283, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3844, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.280969582851044, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3874, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.2984012316645626, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3778, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.30139527047552517, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3994, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 0.24709339303111982, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3827, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.9523809523809526, | |
| "grad_norm": 0.25286199068075177, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.3892, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.0476190476190474, | |
| "grad_norm": 0.3897706495957114, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.5468, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.2720027061358551, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3634, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.238095238095238, | |
| "grad_norm": 0.23340708465551444, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.3698, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.24284971158342084, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3714, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 0.24573930825254336, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.3626, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.523809523809524, | |
| "grad_norm": 0.24228391477999003, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.3656, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.619047619047619, | |
| "grad_norm": 0.2424674393246646, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.3574, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.2260500547448982, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.365, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.809523809523809, | |
| "grad_norm": 0.23262274251274268, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.3644, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.904761904761905, | |
| "grad_norm": 0.24328065113087025, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3682, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.3542176555963666, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.5551, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.095238095238095, | |
| "grad_norm": 0.20725527246794545, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3659, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.190476190476191, | |
| "grad_norm": 0.20871665612226104, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.3638, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.20911819934603798, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.3584, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.380952380952381, | |
| "grad_norm": 0.2119088624352818, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.3661, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.476190476190476, | |
| "grad_norm": 0.20643899236666943, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.3676, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 0.20106120319449272, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.3518, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.20675993672406334, | |
| "learning_rate": 0.0, | |
| "loss": 0.3555, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "step": 70, | |
| "total_flos": 2.3990446016220365e+17, | |
| "train_loss": 0.4939488410949707, | |
| "train_runtime": 2825.0166, | |
| "train_samples_per_second": 2.478, | |
| "train_steps_per_second": 0.025 | |
| } | |
| ], | |
| "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": 2.3990446016220365e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |