| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.75, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09375, | |
| "grad_norm": 6.2971369813800475, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.9121, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 6.426683757053099, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8831, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.28125, | |
| "grad_norm": 6.230761255047608, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.8867, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 2.861526251465193, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.8221, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.46875, | |
| "grad_norm": 4.0110820700217475, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.8055, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "grad_norm": 4.125582564760052, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7486, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.65625, | |
| "grad_norm": 4.429276932299188, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6956, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 3.3384988715157498, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6999, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.84375, | |
| "grad_norm": 2.987114619949788, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6856, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 2.5539980021315336, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.6528, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0625, | |
| "grad_norm": 1.7004789341807878, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.6018, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.15625, | |
| "grad_norm": 1.6606768010026876, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.573, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.52451215562104, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.5681, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.34375, | |
| "grad_norm": 0.9112261661419037, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5629, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4375, | |
| "grad_norm": 1.1596949399919652, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5546, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.53125, | |
| "grad_norm": 0.8676813168885777, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5211, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 0.9246947062153277, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5523, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.71875, | |
| "grad_norm": 0.8993603552929885, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.5248, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8125, | |
| "grad_norm": 0.8345017799173164, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.5185, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.90625, | |
| "grad_norm": 0.7438459569159432, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.5191, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.03125, | |
| "grad_norm": 0.7285402180247241, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.4934, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.125, | |
| "grad_norm": 0.8419193005386686, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4242, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.21875, | |
| "grad_norm": 0.7501344122026129, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4344, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.3125, | |
| "grad_norm": 0.705990416240611, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4115, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.40625, | |
| "grad_norm": 0.677393191496278, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4185, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.803301268956542, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.4153, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.59375, | |
| "grad_norm": 0.6406487907027074, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.3954, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6875, | |
| "grad_norm": 0.6296366184036183, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.3975, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.78125, | |
| "grad_norm": 0.575661455304229, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4195, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.875, | |
| "grad_norm": 0.5858494729408503, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.3955, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.96875, | |
| "grad_norm": 19.302811477660804, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.3866, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.09375, | |
| "grad_norm": 0.624574256675076, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.3292, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.1875, | |
| "grad_norm": 0.5725491788678264, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.3189, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.28125, | |
| "grad_norm": 0.5504342531634668, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.35, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.375, | |
| "grad_norm": 0.5877352595570613, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.3355, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.46875, | |
| "grad_norm": 0.4984637304848101, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3017, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.5625, | |
| "grad_norm": 0.6328664130755941, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3321, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.65625, | |
| "grad_norm": 0.5840160851148782, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.3008, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.5316877322566146, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.2986, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84375, | |
| "grad_norm": 0.499968995017785, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3368, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.9375, | |
| "grad_norm": 0.571348224338469, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.326, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.0625, | |
| "grad_norm": 0.5210021488716393, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.2656, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.15625, | |
| "grad_norm": 0.4950710591120534, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.2752, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 0.5027228466097226, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.2542, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.34375, | |
| "grad_norm": 0.47796898490514245, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.2329, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.4375, | |
| "grad_norm": 0.5088957070990913, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.2767, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.53125, | |
| "grad_norm": 0.4992987195956316, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.2762, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.625, | |
| "grad_norm": 0.4518148637088974, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.2401, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.71875, | |
| "grad_norm": 0.46023344098193936, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.2868, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8125, | |
| "grad_norm": 0.4311098929823891, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.225, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.90625, | |
| "grad_norm": 0.42726221341642917, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.2638, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 5.03125, | |
| "grad_norm": 0.4243068546681569, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.2662, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.125, | |
| "grad_norm": 0.38156299601761867, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.246, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.21875, | |
| "grad_norm": 0.36587806151687596, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2262, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.3125, | |
| "grad_norm": 0.3397738625017687, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2056, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.40625, | |
| "grad_norm": 0.34656811267797805, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.1887, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 0.42551568420004143, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2659, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.59375, | |
| "grad_norm": 0.3501389155228347, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.208, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.6875, | |
| "grad_norm": 0.3790365762890537, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2516, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.78125, | |
| "grad_norm": 0.3505866951353783, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2167, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.875, | |
| "grad_norm": 0.3319477205305615, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2124, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.96875, | |
| "grad_norm": 0.3037992474598154, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2438, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.09375, | |
| "grad_norm": 0.32220842141706346, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.1577, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.1875, | |
| "grad_norm": 0.3079568033955935, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.2637, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.28125, | |
| "grad_norm": 0.310058273869499, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.1791, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.375, | |
| "grad_norm": 0.295574136323341, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2346, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.46875, | |
| "grad_norm": 0.2715628678044081, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2143, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5625, | |
| "grad_norm": 0.2971535951847911, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.1791, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.65625, | |
| "grad_norm": 0.2859988355872997, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.215, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 0.2898256315037482, | |
| "learning_rate": 0.0, | |
| "loss": 0.24, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "step": 70, | |
| "total_flos": 7.568321672246067e+16, | |
| "train_loss": 0.3988671153783798, | |
| "train_runtime": 1546.9539, | |
| "train_samples_per_second": 4.525, | |
| "train_steps_per_second": 0.045 | |
| } | |
| ], | |
| "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": 7.568321672246067e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |