| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 660, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.045454545454545456, | |
| "grad_norm": 11.08301830291748, | |
| "learning_rate": 4.9242424242424245e-05, | |
| "loss": 1.3636, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 6.645240306854248, | |
| "learning_rate": 4.848484848484849e-05, | |
| "loss": 0.948, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13636363636363635, | |
| "grad_norm": 25.287090301513672, | |
| "learning_rate": 4.772727272727273e-05, | |
| "loss": 0.9232, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 10.437969207763672, | |
| "learning_rate": 4.696969696969697e-05, | |
| "loss": 0.7862, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "grad_norm": 4.972208499908447, | |
| "learning_rate": 4.621212121212121e-05, | |
| "loss": 0.7418, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "grad_norm": 16.393909454345703, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.6632, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3181818181818182, | |
| "grad_norm": 11.954917907714844, | |
| "learning_rate": 4.46969696969697e-05, | |
| "loss": 0.7647, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 6.628509044647217, | |
| "learning_rate": 4.3939393939393944e-05, | |
| "loss": 0.8068, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4090909090909091, | |
| "grad_norm": 3.830747365951538, | |
| "learning_rate": 4.318181818181819e-05, | |
| "loss": 0.8173, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.45454545454545453, | |
| "grad_norm": 9.427088737487793, | |
| "learning_rate": 4.242424242424243e-05, | |
| "loss": 0.5975, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 15.573807716369629, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.76, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 4.234420299530029, | |
| "learning_rate": 4.0909090909090915e-05, | |
| "loss": 0.5861, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5909090909090909, | |
| "grad_norm": 10.21924114227295, | |
| "learning_rate": 4.015151515151515e-05, | |
| "loss": 0.6307, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6363636363636364, | |
| "grad_norm": 15.529552459716797, | |
| "learning_rate": 3.939393939393939e-05, | |
| "loss": 0.5305, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6818181818181818, | |
| "grad_norm": 15.605684280395508, | |
| "learning_rate": 3.8636363636363636e-05, | |
| "loss": 0.517, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 4.426260471343994, | |
| "learning_rate": 3.787878787878788e-05, | |
| "loss": 0.6307, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.7727272727272727, | |
| "grad_norm": 6.0872673988342285, | |
| "learning_rate": 3.712121212121212e-05, | |
| "loss": 0.6244, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.8181818181818182, | |
| "grad_norm": 8.042257308959961, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.657, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8636363636363636, | |
| "grad_norm": 5.940113067626953, | |
| "learning_rate": 3.560606060606061e-05, | |
| "loss": 0.5958, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 8.203106880187988, | |
| "learning_rate": 3.484848484848485e-05, | |
| "loss": 0.4277, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9545454545454546, | |
| "grad_norm": 6.048317909240723, | |
| "learning_rate": 3.409090909090909e-05, | |
| "loss": 0.5817, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 10.408190727233887, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.4926, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8351063829787234, | |
| "eval_accuracy_label_communication_issue": 0.0, | |
| "eval_accuracy_label_general_query": 0.0, | |
| "eval_accuracy_label_other": 0.847887323943662, | |
| "eval_accuracy_label_praise": 0.7315436241610739, | |
| "eval_accuracy_label_service_issue": 0.9780346820809248, | |
| "eval_f1": 0.7977035566821362, | |
| "eval_loss": 0.476932555437088, | |
| "eval_precision": 0.7653675138091269, | |
| "eval_recall": 0.8351063829787234, | |
| "eval_runtime": 47.9926, | |
| "eval_samples_per_second": 31.338, | |
| "eval_steps_per_second": 1.959, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.0454545454545454, | |
| "grad_norm": 5.125838756561279, | |
| "learning_rate": 3.257575757575758e-05, | |
| "loss": 0.46, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 6.349830150604248, | |
| "learning_rate": 3.181818181818182e-05, | |
| "loss": 0.4112, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.1363636363636362, | |
| "grad_norm": 19.836454391479492, | |
| "learning_rate": 3.106060606060606e-05, | |
| "loss": 0.3519, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1818181818181819, | |
| "grad_norm": 7.154183864593506, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 0.3793, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.2272727272727273, | |
| "grad_norm": 9.307719230651855, | |
| "learning_rate": 2.954545454545455e-05, | |
| "loss": 0.5391, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 27.64773941040039, | |
| "learning_rate": 2.878787878787879e-05, | |
| "loss": 0.4752, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.3181818181818181, | |
| "grad_norm": 13.110358238220215, | |
| "learning_rate": 2.803030303030303e-05, | |
| "loss": 0.5639, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 10.741105079650879, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.3713, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.4090909090909092, | |
| "grad_norm": 6.00734806060791, | |
| "learning_rate": 2.6515151515151516e-05, | |
| "loss": 0.3944, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 17.730316162109375, | |
| "learning_rate": 2.575757575757576e-05, | |
| "loss": 0.45, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 13.08516788482666, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4115, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.5454545454545454, | |
| "grad_norm": 10.093771934509277, | |
| "learning_rate": 2.4242424242424244e-05, | |
| "loss": 0.2893, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.5909090909090908, | |
| "grad_norm": 10.081781387329102, | |
| "learning_rate": 2.3484848484848487e-05, | |
| "loss": 0.277, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 17.387615203857422, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.5264, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.6818181818181817, | |
| "grad_norm": 9.769563674926758, | |
| "learning_rate": 2.1969696969696972e-05, | |
| "loss": 0.5075, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.7272727272727273, | |
| "grad_norm": 9.551799774169922, | |
| "learning_rate": 2.1212121212121215e-05, | |
| "loss": 0.4975, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.7727272727272727, | |
| "grad_norm": 3.705615758895874, | |
| "learning_rate": 2.0454545454545457e-05, | |
| "loss": 0.3067, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 15.211760520935059, | |
| "learning_rate": 1.9696969696969697e-05, | |
| "loss": 0.4204, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8636363636363638, | |
| "grad_norm": 11.120340347290039, | |
| "learning_rate": 1.893939393939394e-05, | |
| "loss": 0.4252, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.9090909090909092, | |
| "grad_norm": 24.251697540283203, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.2903, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.9545454545454546, | |
| "grad_norm": 6.106864929199219, | |
| "learning_rate": 1.7424242424242425e-05, | |
| "loss": 0.3099, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 497.832275390625, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.4722, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8138297872340425, | |
| "eval_accuracy_label_communication_issue": 0.5918367346938775, | |
| "eval_accuracy_label_general_query": 0.2972972972972973, | |
| "eval_accuracy_label_other": 0.8591549295774648, | |
| "eval_accuracy_label_praise": 0.8523489932885906, | |
| "eval_accuracy_label_service_issue": 0.8358381502890173, | |
| "eval_f1": 0.833458233857523, | |
| "eval_loss": 0.4489530324935913, | |
| "eval_precision": 0.8684601912444823, | |
| "eval_recall": 0.8138297872340425, | |
| "eval_runtime": 48.2944, | |
| "eval_samples_per_second": 31.142, | |
| "eval_steps_per_second": 1.946, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.0454545454545454, | |
| "grad_norm": 4.993196964263916, | |
| "learning_rate": 1.590909090909091e-05, | |
| "loss": 0.3753, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.090909090909091, | |
| "grad_norm": 10.294816017150879, | |
| "learning_rate": 1.5151515151515153e-05, | |
| "loss": 0.227, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.1363636363636362, | |
| "grad_norm": 11.7720365524292, | |
| "learning_rate": 1.4393939393939396e-05, | |
| "loss": 0.3358, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 4.280463695526123, | |
| "learning_rate": 1.3636363636363637e-05, | |
| "loss": 0.2411, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.227272727272727, | |
| "grad_norm": 5.480221271514893, | |
| "learning_rate": 1.287878787878788e-05, | |
| "loss": 0.2354, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.2727272727272725, | |
| "grad_norm": 11.005305290222168, | |
| "learning_rate": 1.2121212121212122e-05, | |
| "loss": 0.3645, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.3181818181818183, | |
| "grad_norm": 8.943399429321289, | |
| "learning_rate": 1.1363636363636365e-05, | |
| "loss": 0.2661, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 16.569984436035156, | |
| "learning_rate": 1.0606060606060607e-05, | |
| "loss": 0.3468, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.409090909090909, | |
| "grad_norm": 16.938270568847656, | |
| "learning_rate": 9.848484848484848e-06, | |
| "loss": 0.2875, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.4545454545454546, | |
| "grad_norm": 16.039297103881836, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.2259, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 8.718212127685547, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.2414, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 5.3960137367248535, | |
| "learning_rate": 7.5757575757575764e-06, | |
| "loss": 0.1906, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.590909090909091, | |
| "grad_norm": 12.368656158447266, | |
| "learning_rate": 6.818181818181818e-06, | |
| "loss": 0.292, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.6363636363636362, | |
| "grad_norm": 7.215673923492432, | |
| "learning_rate": 6.060606060606061e-06, | |
| "loss": 0.2161, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.6818181818181817, | |
| "grad_norm": 10.76251220703125, | |
| "learning_rate": 5.303030303030304e-06, | |
| "loss": 0.2502, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 3.9787635803222656, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 0.3901, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.7727272727272725, | |
| "grad_norm": 17.325733184814453, | |
| "learning_rate": 3.7878787878787882e-06, | |
| "loss": 0.318, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.8181818181818183, | |
| "grad_norm": 2.0341148376464844, | |
| "learning_rate": 3.0303030303030305e-06, | |
| "loss": 0.2546, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.8636363636363638, | |
| "grad_norm": 4.3380279541015625, | |
| "learning_rate": 2.2727272727272728e-06, | |
| "loss": 0.3142, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 57.76769256591797, | |
| "learning_rate": 1.5151515151515152e-06, | |
| "loss": 0.3151, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.9545454545454546, | |
| "grad_norm": 13.915431022644043, | |
| "learning_rate": 7.575757575757576e-07, | |
| "loss": 0.2874, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 11.066320419311523, | |
| "learning_rate": 0.0, | |
| "loss": 0.1949, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8650265957446809, | |
| "eval_accuracy_label_communication_issue": 0.3877551020408163, | |
| "eval_accuracy_label_general_query": 0.5135135135135135, | |
| "eval_accuracy_label_other": 0.8535211267605634, | |
| "eval_accuracy_label_praise": 0.7986577181208053, | |
| "eval_accuracy_label_service_issue": 0.9502890173410404, | |
| "eval_f1": 0.8646844293958481, | |
| "eval_loss": 0.4295935034751892, | |
| "eval_precision": 0.8662149933902713, | |
| "eval_recall": 0.8650265957446809, | |
| "eval_runtime": 48.0636, | |
| "eval_samples_per_second": 31.292, | |
| "eval_steps_per_second": 1.956, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 660, | |
| "total_flos": 2769055331291136.0, | |
| "train_loss": 0.4658627540776224, | |
| "train_runtime": 1450.7799, | |
| "train_samples_per_second": 7.254, | |
| "train_steps_per_second": 0.455 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8650265957446809, | |
| "eval_accuracy_label_communication_issue": 0.3877551020408163, | |
| "eval_accuracy_label_general_query": 0.5135135135135135, | |
| "eval_accuracy_label_other": 0.8535211267605634, | |
| "eval_accuracy_label_praise": 0.7986577181208053, | |
| "eval_accuracy_label_service_issue": 0.9502890173410404, | |
| "eval_f1": 0.8646844293958481, | |
| "eval_loss": 0.4295935034751892, | |
| "eval_precision": 0.8662149933902713, | |
| "eval_recall": 0.8650265957446809, | |
| "eval_runtime": 43.1294, | |
| "eval_samples_per_second": 34.872, | |
| "eval_steps_per_second": 2.179, | |
| "step": 660 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 660, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 2769055331291136.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |