| { | |
| "best_metric": 0.5809483528137207, | |
| "best_model_checkpoint": "neurips-bert-combined1/checkpoint-100", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 3.577862501144409, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.6737, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 5.035987377166748, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.6841, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 3.1154258251190186, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7756, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 10.06010627746582, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.7759, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 7.218173503875732, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.7774, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 9.524176597595215, | |
| "learning_rate": 4e-05, | |
| "loss": 0.74, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 3.1684250831604004, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.6328, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 9.243531227111816, | |
| "learning_rate": 4.962962962962963e-05, | |
| "loss": 0.6749, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 15.164690971374512, | |
| "learning_rate": 4.888888888888889e-05, | |
| "loss": 0.7968, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 4.498668193817139, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.7803, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 10.785091400146484, | |
| "learning_rate": 4.740740740740741e-05, | |
| "loss": 0.6634, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.868218421936035, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.6974, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 6.348710060119629, | |
| "learning_rate": 4.592592592592593e-05, | |
| "loss": 0.6492, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 6.30601692199707, | |
| "learning_rate": 4.518518518518519e-05, | |
| "loss": 0.7361, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 6.230818748474121, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.6889, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 7.42860221862793, | |
| "learning_rate": 4.3703703703703705e-05, | |
| "loss": 0.6431, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 2.7333102226257324, | |
| "learning_rate": 4.296296296296296e-05, | |
| "loss": 0.7148, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 3.3061561584472656, | |
| "learning_rate": 4.222222222222222e-05, | |
| "loss": 0.6978, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 3.080080032348633, | |
| "learning_rate": 4.148148148148148e-05, | |
| "loss": 0.7332, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 2.468094825744629, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.6837, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 4.740555286407471, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7024, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 4.630310535430908, | |
| "learning_rate": 3.925925925925926e-05, | |
| "loss": 0.6951, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 2.2872023582458496, | |
| "learning_rate": 3.851851851851852e-05, | |
| "loss": 0.6575, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.9264609813690186, | |
| "learning_rate": 3.777777777777778e-05, | |
| "loss": 0.6714, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.6678338050842285, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.6892, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.51, | |
| "eval_auc": 0.7687999999999999, | |
| "eval_f1": 0.6711409395973155, | |
| "eval_loss": 0.6739338636398315, | |
| "eval_precision": 0.5050505050505051, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 16.1008, | |
| "eval_samples_per_second": 6.211, | |
| "eval_steps_per_second": 0.435, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 4.582466125488281, | |
| "learning_rate": 3.62962962962963e-05, | |
| "loss": 0.6436, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 6.386196613311768, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 0.6901, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 3.274430751800537, | |
| "learning_rate": 3.481481481481482e-05, | |
| "loss": 0.6233, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 8.776232719421387, | |
| "learning_rate": 3.4074074074074077e-05, | |
| "loss": 0.6726, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 8.833818435668945, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6303, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 7.453307628631592, | |
| "learning_rate": 3.25925925925926e-05, | |
| "loss": 0.6011, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 2.839207887649536, | |
| "learning_rate": 3.185185185185185e-05, | |
| "loss": 0.6144, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 5.08197546005249, | |
| "learning_rate": 3.111111111111111e-05, | |
| "loss": 0.716, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 7.713846206665039, | |
| "learning_rate": 3.037037037037037e-05, | |
| "loss": 0.7857, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 5.370303153991699, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.5537, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 4.6746296882629395, | |
| "learning_rate": 2.8888888888888888e-05, | |
| "loss": 0.5636, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 5.238672733306885, | |
| "learning_rate": 2.814814814814815e-05, | |
| "loss": 0.5311, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 4.931227207183838, | |
| "learning_rate": 2.7407407407407408e-05, | |
| "loss": 0.5448, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 13.78615665435791, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.6089, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 4.783405303955078, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.422, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 10.756999969482422, | |
| "learning_rate": 2.5185185185185183e-05, | |
| "loss": 0.5627, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 8.377484321594238, | |
| "learning_rate": 2.4444444444444445e-05, | |
| "loss": 0.6638, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 6.980154991149902, | |
| "learning_rate": 2.3703703703703707e-05, | |
| "loss": 0.7009, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 5.8498334884643555, | |
| "learning_rate": 2.2962962962962965e-05, | |
| "loss": 0.419, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 7.8378520011901855, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.4252, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 8.54397201538086, | |
| "learning_rate": 2.148148148148148e-05, | |
| "loss": 0.425, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 9.064745903015137, | |
| "learning_rate": 2.074074074074074e-05, | |
| "loss": 0.5294, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 10.862845420837402, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9176, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 6.8356475830078125, | |
| "learning_rate": 1.925925925925926e-05, | |
| "loss": 0.433, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.982166290283203, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.5614, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.69, | |
| "eval_auc": 0.8136, | |
| "eval_f1": 0.6265060240963856, | |
| "eval_loss": 0.5809483528137207, | |
| "eval_precision": 0.7878787878787878, | |
| "eval_recall": 0.52, | |
| "eval_runtime": 16.0696, | |
| "eval_samples_per_second": 6.223, | |
| "eval_steps_per_second": 0.436, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 52622211072000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |