| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.72, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 6.791108718527205, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8588, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.291620895457488, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8102, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.882228123282847, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.8042, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.489955345615572, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7318, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.109257190087163, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7388, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.482235815817922, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7484, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.4838020370242315, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7025, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.7392275828656847, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6754, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.5809715159858264, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6494, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.8194394409280001, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.6309, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.415067068387322, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.6093, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.2014071574880054, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5916, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 0.97213006379183, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.5654, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.8408233195513488, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5616, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.9210467525075646, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5602, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.001703006647715, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5547, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.8511882007854552, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5325, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.6986479458320555, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.5347, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.8421396808888325, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.5559, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.5281889561462852, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.5395, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.6158652039860077, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.4874, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.628781880933258, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4879, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.572331480738609, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4958, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.5168530767148276, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4749, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.43558013365738246, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4835, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.5686783515236928, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.4707, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.39627117444999294, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4626, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.4287914739246222, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.4775, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.41483025696366277, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4518, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.35823375008948694, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.4686, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.40009913605468717, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.4525, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.46238124086181903, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.4451, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.3741314023436948, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4244, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.3766781756497918, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4224, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.4367713378320135, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4382, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.33645868882466423, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.4062, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.3882908597215589, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4235, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.40715401106433835, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4078, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.30830595234716873, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4089, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.3561575950439687, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.4156, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.34558389969215647, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.4083, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.36055013154224563, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.4079, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.36795583294374046, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3906, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.34722724689514955, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3828, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.3451265264183233, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3872, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.3308279101068478, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3794, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.28099871135493887, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3668, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.30541727373987765, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3879, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.2912879348839749, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3708, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.27711903814353656, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3746, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.2939224460483177, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3763, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.26240736489158795, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.365, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.2792593968568827, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.3611, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.2787174792413667, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3583, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.23898339124790235, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.3525, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.23812230919637287, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3522, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.24768581490210512, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.3475, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.2698417186422043, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.366, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.24536740331150125, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.3744, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.2546868986490743, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.3569, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.26956097263120443, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.3425, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.23202697942341147, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.355, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.2347711448346704, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.3625, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.22072609700706103, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3508, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.22251268175082936, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.3509, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.21130340134686762, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.3488, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.22780359143039658, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.342, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.20854010641064077, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.3575, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.22366571570844085, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.3383, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.2345483473888077, | |
| "learning_rate": 0.0, | |
| "loss": 0.3466, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.6855272132431053e+17, | |
| "train_loss": 0.4703236562865121, | |
| "train_runtime": 9432.5268, | |
| "train_samples_per_second": 0.742, | |
| "train_steps_per_second": 0.007 | |
| } | |
| ], | |
| "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": 2.6855272132431053e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |