| { | |
| "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": 6.3835496455762595, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0398, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 6.420129641478437, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0271, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 5.86199569087258, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0152, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 2.847208659782771, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9629, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 4.149211183095472, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9324, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 4.3335284308553765, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9598, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 3.923877630826818, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 2.87431451955724, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.876, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.2070673394156843, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.813, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 1.55995519407377, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.809, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 1.9814752720945792, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.1679, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.468418319969285, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7472, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 1.2526302454828715, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.8338, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 1.010635207662411, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7216, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.9569830783855963, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.6473, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 1.0649627595809863, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7563, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.9037004536927085, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7196, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.8824441627071188, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7322, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.8927603027556651, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.6686, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.7272331900746442, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.701, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.3802328738248533, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.041, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.6665509261324393, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6501, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.7645543670951164, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6355, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.5730034173467009, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6351, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.832612367469944, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6342, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.7167334827479427, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6331, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.7250789354437035, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6235, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.6005936144611025, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6148, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.6292701633425876, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6164, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.6574641818708501, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6188, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.9523809523809526, | |
| "grad_norm": 0.508099635108432, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5996, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.0476190476190474, | |
| "grad_norm": 0.8772575025435417, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.8926, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.5980483723809792, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5919, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.238095238095238, | |
| "grad_norm": 0.4347928512135361, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5647, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.5282006445796072, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5491, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.465381072741314, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5702, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.5238095238095237, | |
| "grad_norm": 0.5113565160914203, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.545, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.619047619047619, | |
| "grad_norm": 0.4647823111141158, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5744, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.5410371675364731, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.566, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.48131684100360583, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5459, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.9047619047619047, | |
| "grad_norm": 0.45651141990895905, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5178, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.8557059082231967, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.8304, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.095238095238095, | |
| "grad_norm": 0.42130779879062297, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5205, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 0.4817432491567549, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5306, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.45846988531366556, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5134, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 0.4139250704681888, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5149, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.476190476190476, | |
| "grad_norm": 0.42768609949080866, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5052, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.4582631847562383, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5045, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.37440070818557053, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5057, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.3551666569194031, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.4984, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 0.40842265499084, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5081, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.9523809523809526, | |
| "grad_norm": 0.41315419352783, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5089, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.0476190476190474, | |
| "grad_norm": 0.5864381016839928, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.7467, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.35124810711264276, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.4848, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.238095238095238, | |
| "grad_norm": 0.3785613322148478, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4922, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.3616314732719367, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4924, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 0.32014898220404536, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4684, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.523809523809524, | |
| "grad_norm": 0.34955825639304905, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.5142, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.619047619047619, | |
| "grad_norm": 0.33909372988234115, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4773, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.2774714670635031, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.4311, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.809523809523809, | |
| "grad_norm": 0.31388010077200085, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.5154, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.904761904761905, | |
| "grad_norm": 0.2772224556215916, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.4438, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.5030320291399264, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.7236, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.095238095238095, | |
| "grad_norm": 0.2874479592227752, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4686, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.190476190476191, | |
| "grad_norm": 0.2807844638880914, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.479, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.2585873292998146, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4736, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.380952380952381, | |
| "grad_norm": 0.28336888521861414, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4715, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.476190476190476, | |
| "grad_norm": 0.27619307735241255, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4671, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 0.2550861636321908, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4859, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.2729279640681754, | |
| "learning_rate": 0.0, | |
| "loss": 0.474, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "step": 70, | |
| "total_flos": 1.9474891998337434e+17, | |
| "train_loss": 0.6471517362764904, | |
| "train_runtime": 3100.5605, | |
| "train_samples_per_second": 2.258, | |
| "train_steps_per_second": 0.023 | |
| } | |
| ], | |
| "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": 1.9474891998337434e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |