| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.152165276122368, |
| "eval_steps": 50, |
| "global_step": 725, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.02, |
| "learning_rate": 3e-05, |
| "loss": 0.9927, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.03, |
| "learning_rate": 3e-05, |
| "loss": 0.1794, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.05, |
| "learning_rate": 3e-05, |
| "loss": 0.1202, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.06, |
| "learning_rate": 3e-05, |
| "loss": 0.0955, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.08, |
| "learning_rate": 3e-05, |
| "loss": 0.0886, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_accuracy": 0.5773584905660377, |
| "eval_f1_macro": 0.32220178110380887, |
| "eval_f1_micro": 0.5773584905660377, |
| "eval_loss": 0.10824274271726608, |
| "eval_precision_macro": 0.3987863781703509, |
| "eval_precision_micro": 0.5773584905660377, |
| "eval_recall_macro": 0.31235484191508905, |
| "eval_recall_micro": 0.5773584905660377, |
| "eval_runtime": 68.6809, |
| "eval_samples_per_second": 15.434, |
| "eval_steps_per_second": 3.858, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.1, |
| "learning_rate": 3e-05, |
| "loss": 0.2261, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.11, |
| "learning_rate": 3e-05, |
| "loss": 0.0728, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.13, |
| "learning_rate": 3e-05, |
| "loss": 0.0659, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.14, |
| "learning_rate": 3e-05, |
| "loss": 0.051, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.16, |
| "learning_rate": 3e-05, |
| "loss": 0.0572, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_accuracy": 0.5877358490566038, |
| "eval_f1_macro": 0.379683051721915, |
| "eval_f1_micro": 0.5877358490566038, |
| "eval_loss": 0.08316469192504883, |
| "eval_precision_macro": 0.4716361677310224, |
| "eval_precision_micro": 0.5877358490566038, |
| "eval_recall_macro": 0.3681154625887901, |
| "eval_recall_micro": 0.5877358490566038, |
| "eval_runtime": 68.5728, |
| "eval_samples_per_second": 15.458, |
| "eval_steps_per_second": 3.865, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17, |
| "learning_rate": 3e-05, |
| "loss": 0.1229, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.19, |
| "learning_rate": 3e-05, |
| "loss": 0.0561, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.21, |
| "learning_rate": 3e-05, |
| "loss": 0.0549, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.22, |
| "learning_rate": 3e-05, |
| "loss": 0.0562, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.24, |
| "learning_rate": 3e-05, |
| "loss": 0.0496, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.24, |
| "eval_accuracy": 0.7311320754716981, |
| "eval_f1_macro": 0.5702888507555568, |
| "eval_f1_micro": 0.7311320754716981, |
| "eval_loss": 0.05246850848197937, |
| "eval_precision_macro": 0.5911132598584014, |
| "eval_precision_micro": 0.7311320754716981, |
| "eval_recall_macro": 0.5746634992387409, |
| "eval_recall_micro": 0.7311320754716981, |
| "eval_runtime": 68.5467, |
| "eval_samples_per_second": 15.464, |
| "eval_steps_per_second": 3.866, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.25, |
| "learning_rate": 3e-05, |
| "loss": 0.067, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.27, |
| "learning_rate": 3e-05, |
| "loss": 0.0551, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.29, |
| "learning_rate": 3e-05, |
| "loss": 0.0426, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.3, |
| "learning_rate": 3e-05, |
| "loss": 0.0475, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.32, |
| "learning_rate": 3e-05, |
| "loss": 0.0541, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_accuracy": 0.7566037735849057, |
| "eval_f1_macro": 0.5584352420606399, |
| "eval_f1_micro": 0.7566037735849057, |
| "eval_loss": 0.04639677330851555, |
| "eval_precision_macro": 0.6151030653662273, |
| "eval_precision_micro": 0.7566037735849057, |
| "eval_recall_macro": 0.5606200967693734, |
| "eval_recall_micro": 0.7566037735849057, |
| "eval_runtime": 68.5714, |
| "eval_samples_per_second": 15.458, |
| "eval_steps_per_second": 3.865, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.33, |
| "learning_rate": 3e-05, |
| "loss": 0.052, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.35, |
| "learning_rate": 3e-05, |
| "loss": 0.0435, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.37, |
| "learning_rate": 3e-05, |
| "loss": 0.0407, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.38, |
| "learning_rate": 3e-05, |
| "loss": 0.0475, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.4, |
| "learning_rate": 3e-05, |
| "loss": 0.0481, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_accuracy": 0.7811320754716982, |
| "eval_f1_macro": 0.6368753336945477, |
| "eval_f1_micro": 0.7811320754716982, |
| "eval_loss": 0.04328591376543045, |
| "eval_precision_macro": 0.6636054486047108, |
| "eval_precision_micro": 0.7811320754716982, |
| "eval_recall_macro": 0.6513962288983521, |
| "eval_recall_micro": 0.7811320754716982, |
| "eval_runtime": 68.5331, |
| "eval_samples_per_second": 15.467, |
| "eval_steps_per_second": 3.867, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.41, |
| "learning_rate": 3e-05, |
| "loss": 0.0533, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.43, |
| "learning_rate": 3e-05, |
| "loss": 0.0433, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.44, |
| "learning_rate": 3e-05, |
| "loss": 0.0432, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.46, |
| "learning_rate": 3e-05, |
| "loss": 0.0466, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.48, |
| "learning_rate": 3e-05, |
| "loss": 0.053, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_accuracy": 0.7632075471698113, |
| "eval_f1_macro": 0.6337788769448006, |
| "eval_f1_micro": 0.7632075471698113, |
| "eval_loss": 0.0451766662299633, |
| "eval_precision_macro": 0.6935549047637881, |
| "eval_precision_micro": 0.7632075471698113, |
| "eval_recall_macro": 0.6461210681607904, |
| "eval_recall_micro": 0.7632075471698113, |
| "eval_runtime": 68.6668, |
| "eval_samples_per_second": 15.437, |
| "eval_steps_per_second": 3.859, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.49, |
| "learning_rate": 3e-05, |
| "loss": 0.0455, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.51, |
| "learning_rate": 3e-05, |
| "loss": 0.049, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.52, |
| "learning_rate": 3e-05, |
| "loss": 0.0426, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.54, |
| "learning_rate": 3e-05, |
| "loss": 0.0396, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.56, |
| "learning_rate": 3e-05, |
| "loss": 0.0401, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.56, |
| "eval_accuracy": 0.7943396226415095, |
| "eval_f1_macro": 0.6696815276160976, |
| "eval_f1_micro": 0.7943396226415095, |
| "eval_loss": 0.039866555482149124, |
| "eval_precision_macro": 0.7380690805686413, |
| "eval_precision_micro": 0.7943396226415095, |
| "eval_recall_macro": 0.660366139065656, |
| "eval_recall_micro": 0.7943396226415095, |
| "eval_runtime": 68.673, |
| "eval_samples_per_second": 15.435, |
| "eval_steps_per_second": 3.859, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.57, |
| "learning_rate": 3e-05, |
| "loss": 0.0482, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.59, |
| "learning_rate": 3e-05, |
| "loss": 0.0373, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.6, |
| "learning_rate": 3e-05, |
| "loss": 0.036, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 3e-05, |
| "loss": 0.0428, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.64, |
| "learning_rate": 3e-05, |
| "loss": 0.0509, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_accuracy": 0.8009433962264151, |
| "eval_f1_macro": 0.6501313860427956, |
| "eval_f1_micro": 0.8009433962264151, |
| "eval_loss": 0.03930637985467911, |
| "eval_precision_macro": 0.6546476325081232, |
| "eval_precision_micro": 0.8009433962264151, |
| "eval_recall_macro": 0.6611935636788673, |
| "eval_recall_micro": 0.8009433962264151, |
| "eval_runtime": 68.557, |
| "eval_samples_per_second": 15.462, |
| "eval_steps_per_second": 3.865, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.65, |
| "learning_rate": 3e-05, |
| "loss": 0.0375, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.67, |
| "learning_rate": 3e-05, |
| "loss": 0.041, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.68, |
| "learning_rate": 3e-05, |
| "loss": 0.0416, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.7, |
| "learning_rate": 3e-05, |
| "loss": 0.0396, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.72, |
| "learning_rate": 3e-05, |
| "loss": 0.0474, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_accuracy": 0.8018867924528302, |
| "eval_f1_macro": 0.6864569711826704, |
| "eval_f1_micro": 0.8018867924528302, |
| "eval_loss": 0.04012966528534889, |
| "eval_precision_macro": 0.7255429634365795, |
| "eval_precision_micro": 0.8018867924528302, |
| "eval_recall_macro": 0.6926779368041328, |
| "eval_recall_micro": 0.8018867924528302, |
| "eval_runtime": 68.6193, |
| "eval_samples_per_second": 15.448, |
| "eval_steps_per_second": 3.862, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.73, |
| "learning_rate": 3e-05, |
| "loss": 0.0434, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.75, |
| "learning_rate": 3e-05, |
| "loss": 0.0358, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.76, |
| "learning_rate": 3e-05, |
| "loss": 0.0416, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.78, |
| "learning_rate": 3e-05, |
| "loss": 0.0334, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.79, |
| "learning_rate": 3e-05, |
| "loss": 0.045, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.79, |
| "eval_accuracy": 0.8009433962264151, |
| "eval_f1_macro": 0.6977412107603574, |
| "eval_f1_micro": 0.8009433962264151, |
| "eval_loss": 0.0379195362329483, |
| "eval_precision_macro": 0.7146704097806501, |
| "eval_precision_micro": 0.8009433962264151, |
| "eval_recall_macro": 0.710805016133364, |
| "eval_recall_micro": 0.8009433962264151, |
| "eval_runtime": 68.6344, |
| "eval_samples_per_second": 15.444, |
| "eval_steps_per_second": 3.861, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.81, |
| "learning_rate": 3e-05, |
| "loss": 0.0425, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.83, |
| "learning_rate": 3e-05, |
| "loss": 0.036, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.84, |
| "learning_rate": 3e-05, |
| "loss": 0.0444, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.86, |
| "learning_rate": 3e-05, |
| "loss": 0.0394, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.87, |
| "learning_rate": 3e-05, |
| "loss": 0.0335, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.87, |
| "eval_accuracy": 0.8150943396226416, |
| "eval_f1_macro": 0.7134710025440245, |
| "eval_f1_micro": 0.8150943396226416, |
| "eval_loss": 0.03691105917096138, |
| "eval_precision_macro": 0.7046165923538829, |
| "eval_precision_micro": 0.8150943396226416, |
| "eval_recall_macro": 0.7335435173781253, |
| "eval_recall_micro": 0.8150943396226416, |
| "eval_runtime": 68.5596, |
| "eval_samples_per_second": 15.461, |
| "eval_steps_per_second": 3.865, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.89, |
| "learning_rate": 3e-05, |
| "loss": 0.0421, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.91, |
| "learning_rate": 3e-05, |
| "loss": 0.0407, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.92, |
| "learning_rate": 3e-05, |
| "loss": 0.0429, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.94, |
| "learning_rate": 3e-05, |
| "loss": 0.0378, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.95, |
| "learning_rate": 3e-05, |
| "loss": 0.0429, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.95, |
| "eval_accuracy": 0.7962264150943397, |
| "eval_f1_macro": 0.687832173620461, |
| "eval_f1_micro": 0.7962264150943396, |
| "eval_loss": 0.03668661788105965, |
| "eval_precision_macro": 0.7081030422724828, |
| "eval_precision_micro": 0.7962264150943397, |
| "eval_recall_macro": 0.6958905634881637, |
| "eval_recall_micro": 0.7962264150943397, |
| "eval_runtime": 68.6225, |
| "eval_samples_per_second": 15.447, |
| "eval_steps_per_second": 3.862, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.97, |
| "learning_rate": 3e-05, |
| "loss": 0.0394, |
| "step": 610 |
| }, |
| { |
| "epoch": 0.99, |
| "learning_rate": 3e-05, |
| "loss": 0.0363, |
| "step": 620 |
| }, |
| { |
| "epoch": 1.0, |
| "learning_rate": 3e-05, |
| "loss": 0.0428, |
| "step": 630 |
| }, |
| { |
| "epoch": 1.02, |
| "learning_rate": 3e-05, |
| "loss": 0.0265, |
| "step": 640 |
| }, |
| { |
| "epoch": 1.03, |
| "learning_rate": 3e-05, |
| "loss": 0.0253, |
| "step": 650 |
| }, |
| { |
| "epoch": 1.03, |
| "eval_accuracy": 0.8254716981132075, |
| "eval_f1_macro": 0.7098363543260356, |
| "eval_f1_micro": 0.8254716981132075, |
| "eval_loss": 0.03421418368816376, |
| "eval_precision_macro": 0.7370277877974788, |
| "eval_precision_micro": 0.8254716981132075, |
| "eval_recall_macro": 0.6974501533785277, |
| "eval_recall_micro": 0.8254716981132075, |
| "eval_runtime": 68.5452, |
| "eval_samples_per_second": 15.464, |
| "eval_steps_per_second": 3.866, |
| "step": 650 |
| }, |
| { |
| "epoch": 1.05, |
| "learning_rate": 3e-05, |
| "loss": 0.0243, |
| "step": 660 |
| }, |
| { |
| "epoch": 1.06, |
| "learning_rate": 3e-05, |
| "loss": 0.0313, |
| "step": 670 |
| }, |
| { |
| "epoch": 1.08, |
| "learning_rate": 3e-05, |
| "loss": 0.0285, |
| "step": 680 |
| }, |
| { |
| "epoch": 1.1, |
| "learning_rate": 3e-05, |
| "loss": 0.0262, |
| "step": 690 |
| }, |
| { |
| "epoch": 1.11, |
| "learning_rate": 3e-05, |
| "loss": 0.0311, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.11, |
| "eval_accuracy": 0.8047169811320755, |
| "eval_f1_macro": 0.6661175684975367, |
| "eval_f1_micro": 0.8047169811320755, |
| "eval_loss": 0.0357016883790493, |
| "eval_precision_macro": 0.6994736595477448, |
| "eval_precision_micro": 0.8047169811320755, |
| "eval_recall_macro": 0.660920267471505, |
| "eval_recall_micro": 0.8047169811320755, |
| "eval_runtime": 68.6332, |
| "eval_samples_per_second": 15.444, |
| "eval_steps_per_second": 3.861, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.13, |
| "learning_rate": 3e-05, |
| "loss": 0.0311, |
| "step": 710 |
| }, |
| { |
| "epoch": 1.14, |
| "learning_rate": 3e-05, |
| "loss": 0.0298, |
| "step": 720 |
| }, |
| { |
| "epoch": 1.15, |
| "step": 725, |
| "total_flos": 3.949643257934285e+17, |
| "train_loss": 0.06415341473858932, |
| "train_runtime": 4786.626, |
| "train_samples_per_second": 4.847, |
| "train_steps_per_second": 0.151 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 725, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 250, |
| "total_flos": 3.949643257934285e+17, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
