| { | |
| "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": 6.575792410050601, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.09, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.462288855931108, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0736, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.1634056501988965, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0669, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.6542024192998146, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.955, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.5509642847085585, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9518, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.893730534879448, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9211, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 5.0617509363355, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9174, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 4.108961489722247, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8942, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 3.684313713781509, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.871, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.808528285889541, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8327, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.4980902569372874, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.7748, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.8881538378217393, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.754, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 2.0044229011265826, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7335, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.9975951318377048, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7473, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.3506809918484313, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.723, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.2996812942704945, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7236, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.1589215903772327, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7089, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.939946673665635, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.6917, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 1.0613128033282793, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.8072730473932119, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6333, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.0329981294174058, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.6483, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 1.061832941112662, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.5861, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.84254603321981, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.5568, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.9439953231956922, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.5518, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.8911387938710289, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.547, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.7910311200642371, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5474, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7760439723857149, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5241, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6857460195341345, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5105, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.8247868182127068, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.5272, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.6651903010730649, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5222, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.7250350593623238, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.512, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.7407962610603765, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.4468, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.6917188604365652, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4129, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.7337543432964279, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4644, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.7087671619043305, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4232, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.6587527992465353, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3894, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.6528419479325293, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.378, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.6367255565911163, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4264, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.6235062638465595, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.3715, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.5995746241096439, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.4145, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.5707794045728788, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3809, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.5668495217138063, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.3983, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.5678355168329486, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3303, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.5820170786418272, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3167, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.5193880224143694, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3448, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.6191700089765936, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3111, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.6142849942868789, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3518, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.5324725072099992, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3495, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.5125500472433245, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3333, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.5336201472810086, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3216, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.5081320879401945, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.2864, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.44398471499548386, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.3101, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.4465808797731063, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2669, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.44589463380965705, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3012, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.4238759411984323, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2746, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.468118854428135, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2823, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.4185617499975625, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2678, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.3999243072693133, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.2844, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.41258040895556164, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2796, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.39053171043520285, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2649, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.39663765692828784, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2933, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3808171195715162, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2634, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.38385243932610763, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.2919, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.42085527061274086, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.2577, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.38071896577246717, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2624, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.35262808855503014, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2692, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.32472084139270185, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2438, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.3324211819044101, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.28, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.3271228311499256, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2875, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.36512437584067514, | |
| "learning_rate": 0.0, | |
| "loss": 0.2616, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.1894617306614989e+17, | |
| "train_loss": 0.5070248737931251, | |
| "train_runtime": 8475.7864, | |
| "train_samples_per_second": 0.826, | |
| "train_steps_per_second": 0.008 | |
| } | |
| ], | |
| "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": 1.1894617306614989e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |