| { | |
| "best_metric": 0.8897849462365591, | |
| "best_model_checkpoint": "./results/checkpoint-465", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 465, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.10752688172043011, | |
| "grad_norm": 77.31194305419922, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.0774, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.21505376344086022, | |
| "grad_norm": 21.743606567382812, | |
| "learning_rate": 3e-06, | |
| "loss": 1.205, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3225806451612903, | |
| "grad_norm": 46.53227233886719, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.9877, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.43010752688172044, | |
| "grad_norm": 60.25495147705078, | |
| "learning_rate": 6e-06, | |
| "loss": 0.7626, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5376344086021505, | |
| "grad_norm": 42.291969299316406, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.6559, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6451612903225806, | |
| "grad_norm": 27.09429931640625, | |
| "learning_rate": 9e-06, | |
| "loss": 0.3539, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7526881720430108, | |
| "grad_norm": 29.117910385131836, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.516, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8602150537634409, | |
| "grad_norm": 35.026817321777344, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.5457, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.967741935483871, | |
| "grad_norm": 17.851289749145508, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.5821, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8655913978494624, | |
| "eval_f1": 0.8661634583786485, | |
| "eval_loss": 0.3687117397785187, | |
| "eval_runtime": 1.6159, | |
| "eval_samples_per_second": 230.212, | |
| "eval_steps_per_second": 3.713, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.075268817204301, | |
| "grad_norm": 13.748136520385742, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.307, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1827956989247312, | |
| "grad_norm": 3.2135958671569824, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.4692, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2903225806451613, | |
| "grad_norm": 11.485154151916504, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.3206, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3978494623655915, | |
| "grad_norm": 7.797122478485107, | |
| "learning_rate": 1.95e-05, | |
| "loss": 0.3138, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5053763440860215, | |
| "grad_norm": 6.313217639923096, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.3135, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "grad_norm": 18.319936752319336, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2156, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.7204301075268817, | |
| "grad_norm": 21.939163208007812, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.2749, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.827956989247312, | |
| "grad_norm": 20.2148380279541, | |
| "learning_rate": 2.55e-05, | |
| "loss": 0.4502, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.935483870967742, | |
| "grad_norm": 22.044946670532227, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.364, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8763440860215054, | |
| "eval_f1": 0.8742605847498066, | |
| "eval_loss": 0.34187108278274536, | |
| "eval_runtime": 1.6404, | |
| "eval_samples_per_second": 226.774, | |
| "eval_steps_per_second": 3.658, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 2.043010752688172, | |
| "grad_norm": 1.5786309242248535, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.251, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.150537634408602, | |
| "grad_norm": 0.3100162446498871, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0491, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.258064516129032, | |
| "grad_norm": 10.301804542541504, | |
| "learning_rate": 2.8867924528301887e-05, | |
| "loss": 0.0978, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.3655913978494625, | |
| "grad_norm": 0.5391497015953064, | |
| "learning_rate": 2.7735849056603773e-05, | |
| "loss": 0.2325, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.4731182795698925, | |
| "grad_norm": 33.04327392578125, | |
| "learning_rate": 2.6603773584905663e-05, | |
| "loss": 0.3053, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.5806451612903225, | |
| "grad_norm": 1.8449115753173828, | |
| "learning_rate": 2.547169811320755e-05, | |
| "loss": 0.1469, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.688172043010753, | |
| "grad_norm": 41.86624526977539, | |
| "learning_rate": 2.4339622641509435e-05, | |
| "loss": 0.0985, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.795698924731183, | |
| "grad_norm": 12.837655067443848, | |
| "learning_rate": 2.320754716981132e-05, | |
| "loss": 0.1707, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.903225806451613, | |
| "grad_norm": 7.691096305847168, | |
| "learning_rate": 2.2075471698113208e-05, | |
| "loss": 0.1889, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8548387096774194, | |
| "eval_f1": 0.8514191596326964, | |
| "eval_loss": 0.5420617461204529, | |
| "eval_runtime": 1.6403, | |
| "eval_samples_per_second": 226.793, | |
| "eval_steps_per_second": 3.658, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 3.010752688172043, | |
| "grad_norm": 24.820016860961914, | |
| "learning_rate": 2.0943396226415094e-05, | |
| "loss": 0.2798, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.118279569892473, | |
| "grad_norm": 0.1231377124786377, | |
| "learning_rate": 1.981132075471698e-05, | |
| "loss": 0.1421, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.225806451612903, | |
| "grad_norm": 22.27264404296875, | |
| "learning_rate": 1.8679245283018867e-05, | |
| "loss": 0.1455, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 43.31319046020508, | |
| "learning_rate": 1.7547169811320753e-05, | |
| "loss": 0.1223, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.4408602150537635, | |
| "grad_norm": 0.16742512583732605, | |
| "learning_rate": 1.6415094339622643e-05, | |
| "loss": 0.0214, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.5483870967741935, | |
| "grad_norm": 44.84831619262695, | |
| "learning_rate": 1.528301886792453e-05, | |
| "loss": 0.0278, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.6559139784946235, | |
| "grad_norm": 16.63416862487793, | |
| "learning_rate": 1.4150943396226415e-05, | |
| "loss": 0.0238, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.763440860215054, | |
| "grad_norm": 0.019003387540578842, | |
| "learning_rate": 1.3018867924528303e-05, | |
| "loss": 0.0788, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.870967741935484, | |
| "grad_norm": 0.062045346945524216, | |
| "learning_rate": 1.188679245283019e-05, | |
| "loss": 0.0247, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.978494623655914, | |
| "grad_norm": 36.90345001220703, | |
| "learning_rate": 1.0754716981132076e-05, | |
| "loss": 0.1049, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8844086021505376, | |
| "eval_f1": 0.8851071788381587, | |
| "eval_loss": 0.5284830927848816, | |
| "eval_runtime": 1.6712, | |
| "eval_samples_per_second": 222.589, | |
| "eval_steps_per_second": 3.59, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 4.086021505376344, | |
| "grad_norm": 0.04606785252690315, | |
| "learning_rate": 9.622641509433962e-06, | |
| "loss": 0.0563, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.193548387096774, | |
| "grad_norm": 0.11379247903823853, | |
| "learning_rate": 8.49056603773585e-06, | |
| "loss": 0.0007, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.301075268817204, | |
| "grad_norm": 0.057167768478393555, | |
| "learning_rate": 7.358490566037736e-06, | |
| "loss": 0.0008, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.408602150537634, | |
| "grad_norm": 1.8930367231369019, | |
| "learning_rate": 6.226415094339623e-06, | |
| "loss": 0.0016, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.516129032258064, | |
| "grad_norm": 0.10506568104028702, | |
| "learning_rate": 5.094339622641509e-06, | |
| "loss": 0.0183, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.623655913978495, | |
| "grad_norm": 0.29419633746147156, | |
| "learning_rate": 3.962264150943396e-06, | |
| "loss": 0.0379, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.731182795698925, | |
| "grad_norm": 1.4433554410934448, | |
| "learning_rate": 2.830188679245283e-06, | |
| "loss": 0.0011, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.838709677419355, | |
| "grad_norm": 0.05199088156223297, | |
| "learning_rate": 1.69811320754717e-06, | |
| "loss": 0.0276, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.946236559139785, | |
| "grad_norm": 0.017505839467048645, | |
| "learning_rate": 5.660377358490566e-07, | |
| "loss": 0.0009, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8897849462365591, | |
| "eval_f1": 0.8902518525126154, | |
| "eval_loss": 0.518719494342804, | |
| "eval_runtime": 1.6655, | |
| "eval_samples_per_second": 223.351, | |
| "eval_steps_per_second": 3.602, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 465, | |
| "total_flos": 312883992936900.0, | |
| "train_loss": 0.26607618075144546, | |
| "train_runtime": 161.9779, | |
| "train_samples_per_second": 45.84, | |
| "train_steps_per_second": 2.871 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8897849462365591, | |
| "eval_f1": 0.8902518525126154, | |
| "eval_loss": 0.518719494342804, | |
| "eval_runtime": 1.6415, | |
| "eval_samples_per_second": 226.619, | |
| "eval_steps_per_second": 3.655, | |
| "step": 465 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 465, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 312883992936900.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |