| { | |
| "best_metric": 0.7474747474747475, | |
| "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-13/checkpoint-192", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 192, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 1.2512309551239014, | |
| "learning_rate": 1.034188035217011e-05, | |
| "loss": 0.7013, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 1.0938308238983154, | |
| "learning_rate": 2.068376070434022e-05, | |
| "loss": 0.6709, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 1.029103398323059, | |
| "learning_rate": 3.102564105651033e-05, | |
| "loss": 0.676, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.7623715400695801, | |
| "learning_rate": 4.136752140868044e-05, | |
| "loss": 0.6371, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.6755096912384033, | |
| "learning_rate": 5.170940176085055e-05, | |
| "loss": 0.6148, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.48364195227622986, | |
| "learning_rate": 6.205128211302065e-05, | |
| "loss": 0.609, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 3.724107503890991, | |
| "learning_rate": 7.239316246519077e-05, | |
| "loss": 0.5257, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 1.4992948770523071, | |
| "learning_rate": 8.273504281736088e-05, | |
| "loss": 0.7212, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 4.220996856689453, | |
| "learning_rate": 9.307692316953099e-05, | |
| "loss": 0.6503, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.7372970581054688, | |
| "eval_runtime": 1.384, | |
| "eval_samples_per_second": 46.242, | |
| "eval_steps_per_second": 5.78, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 4.6990180015563965, | |
| "learning_rate": 9.882241225406994e-05, | |
| "loss": 0.6895, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 1.5386626720428467, | |
| "learning_rate": 9.767331443716214e-05, | |
| "loss": 0.6802, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.7925231456756592, | |
| "learning_rate": 9.652421662025435e-05, | |
| "loss": 0.4231, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 1.0101344585418701, | |
| "learning_rate": 9.560493836672811e-05, | |
| "loss": 0.5768, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 1.2472072839736938, | |
| "learning_rate": 9.445584054982033e-05, | |
| "loss": 0.5095, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 3.8222692012786865, | |
| "learning_rate": 9.330674273291253e-05, | |
| "loss": 0.8618, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 2.367830276489258, | |
| "learning_rate": 9.215764491600475e-05, | |
| "loss": 0.6321, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 1.2242908477783203, | |
| "learning_rate": 9.100854709909696e-05, | |
| "loss": 0.6086, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.1424415111541748, | |
| "learning_rate": 8.985944928218916e-05, | |
| "loss": 0.6172, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 1.9207895994186401, | |
| "learning_rate": 8.871035146528138e-05, | |
| "loss": 0.6318, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.6706695556640625, | |
| "eval_runtime": 1.3744, | |
| "eval_samples_per_second": 46.567, | |
| "eval_steps_per_second": 5.821, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.3165736198425293, | |
| "learning_rate": 8.756125364837358e-05, | |
| "loss": 0.6359, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 1.2420493364334106, | |
| "learning_rate": 8.64121558314658e-05, | |
| "loss": 0.5142, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 1.2290922403335571, | |
| "learning_rate": 8.526305801455801e-05, | |
| "loss": 0.643, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.1532477140426636, | |
| "learning_rate": 8.434377976103179e-05, | |
| "loss": 0.6774, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 6.360584259033203, | |
| "learning_rate": 8.319468194412399e-05, | |
| "loss": 0.5839, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 7.5169677734375, | |
| "learning_rate": 8.204558412721619e-05, | |
| "loss": 0.5138, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "grad_norm": 14.515143394470215, | |
| "learning_rate": 8.089648631030841e-05, | |
| "loss": 0.6207, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 2.682206153869629, | |
| "learning_rate": 7.974738849340061e-05, | |
| "loss": 0.4247, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 5.204345226287842, | |
| "learning_rate": 7.88281102398744e-05, | |
| "loss": 0.3698, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.9202766418457031, | |
| "eval_runtime": 1.4182, | |
| "eval_samples_per_second": 45.129, | |
| "eval_steps_per_second": 5.641, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 6.481989860534668, | |
| "learning_rate": 7.76790124229666e-05, | |
| "loss": 0.6196, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": Infinity, | |
| "learning_rate": 7.675973416944037e-05, | |
| "loss": 0.4192, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 2.9529616832733154, | |
| "learning_rate": 7.561063635253257e-05, | |
| "loss": 0.2996, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "grad_norm": 13.315085411071777, | |
| "learning_rate": 7.469135809900635e-05, | |
| "loss": 0.6014, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 3.830242156982422, | |
| "learning_rate": 7.354226028209855e-05, | |
| "loss": 0.2117, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "grad_norm": 26.10947608947754, | |
| "learning_rate": 7.239316246519077e-05, | |
| "loss": 0.2918, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "grad_norm": 55.73325729370117, | |
| "learning_rate": 7.124406464828297e-05, | |
| "loss": 0.775, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 1.9401394128799438, | |
| "learning_rate": 7.009496683137518e-05, | |
| "loss": 0.4163, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "grad_norm": 0.818503201007843, | |
| "learning_rate": 6.89458690144674e-05, | |
| "loss": 0.3559, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 6.375278472900391, | |
| "learning_rate": 6.77967711975596e-05, | |
| "loss": 0.5548, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.7474747474747475, | |
| "eval_loss": 1.170898675918579, | |
| "eval_runtime": 1.3939, | |
| "eval_samples_per_second": 45.913, | |
| "eval_steps_per_second": 5.739, | |
| "step": 192 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2891755054954176.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 9.928205138083305e-05, | |
| "per_device_train_batch_size": 4 | |
| } | |
| } | |