| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.0131800174713135, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5623, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7167919799498746, | |
| "eval_f1": 0.5794790005316321, | |
| "eval_loss": 0.5053456425666809, | |
| "eval_precision": 0.6409822866344606, | |
| "eval_recall": 0.5796053827968721, | |
| "eval_runtime": 5.6071, | |
| "eval_samples_per_second": 71.159, | |
| "eval_steps_per_second": 8.917, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.634490966796875, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.518, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7293233082706767, | |
| "eval_f1": 0.599784530797236, | |
| "eval_loss": 0.4860531687736511, | |
| "eval_precision": 0.6673625792811839, | |
| "eval_recall": 0.5959719949081652, | |
| "eval_runtime": 5.7755, | |
| "eval_samples_per_second": 69.085, | |
| "eval_steps_per_second": 8.657, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.959808111190796, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4835, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7694235588972431, | |
| "eval_f1": 0.7145034843205575, | |
| "eval_loss": 0.45518842339515686, | |
| "eval_precision": 0.7210824478299833, | |
| "eval_recall": 0.7093562465902892, | |
| "eval_runtime": 5.2584, | |
| "eval_samples_per_second": 75.878, | |
| "eval_steps_per_second": 9.509, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 4.635169506072998, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4497, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7944862155388471, | |
| "eval_f1": 0.7520912893253319, | |
| "eval_loss": 0.4223441481590271, | |
| "eval_precision": 0.7520912893253319, | |
| "eval_recall": 0.7520912893253319, | |
| "eval_runtime": 5.0487, | |
| "eval_samples_per_second": 79.03, | |
| "eval_steps_per_second": 9.903, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 8.219679832458496, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4266, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8170426065162907, | |
| "eval_f1": 0.7740779522978476, | |
| "eval_loss": 0.399569034576416, | |
| "eval_precision": 0.7814051164566629, | |
| "eval_recall": 0.7680487361338425, | |
| "eval_runtime": 5.0767, | |
| "eval_samples_per_second": 78.595, | |
| "eval_steps_per_second": 9.849, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.150725841522217, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3907, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_f1": 0.784453781512605, | |
| "eval_loss": 0.3830115497112274, | |
| "eval_precision": 0.7818241274748796, | |
| "eval_recall": 0.787324968176032, | |
| "eval_runtime": 5.0718, | |
| "eval_samples_per_second": 78.67, | |
| "eval_steps_per_second": 9.858, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.297985076904297, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3742, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.798423147581139, | |
| "eval_loss": 0.3684135675430298, | |
| "eval_precision": 0.8016430472182685, | |
| "eval_recall": 0.7954628114202582, | |
| "eval_runtime": 5.0743, | |
| "eval_samples_per_second": 78.632, | |
| "eval_steps_per_second": 9.854, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 8.395323753356934, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3616, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_f1": 0.7967966933608887, | |
| "eval_loss": 0.3719731867313385, | |
| "eval_precision": 0.7902444649446494, | |
| "eval_recall": 0.8051463902527732, | |
| "eval_runtime": 5.0484, | |
| "eval_samples_per_second": 79.035, | |
| "eval_steps_per_second": 9.904, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 3.748974561691284, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3294, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8076965854743632, | |
| "eval_loss": 0.36888691782951355, | |
| "eval_precision": 0.8018925518925519, | |
| "eval_recall": 0.8147390434624477, | |
| "eval_runtime": 5.0543, | |
| "eval_samples_per_second": 78.943, | |
| "eval_steps_per_second": 9.893, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 9.309541702270508, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3207, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.8110907261644079, | |
| "eval_loss": 0.36315786838531494, | |
| "eval_precision": 0.8046983557202408, | |
| "eval_recall": 0.819012547735952, | |
| "eval_runtime": 5.0709, | |
| "eval_samples_per_second": 78.684, | |
| "eval_steps_per_second": 9.86, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.2568168640136719, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.3214, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8085765951950401, | |
| "eval_loss": 0.3577338457107544, | |
| "eval_precision": 0.8017470018450185, | |
| "eval_recall": 0.817239498090562, | |
| "eval_runtime": 5.1071, | |
| "eval_samples_per_second": 78.126, | |
| "eval_steps_per_second": 9.79, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.915198802947998, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3167, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.8119476846942383, | |
| "eval_loss": 0.36069995164871216, | |
| "eval_precision": 0.8045650301464256, | |
| "eval_recall": 0.8215130023640662, | |
| "eval_runtime": 5.0598, | |
| "eval_samples_per_second": 78.857, | |
| "eval_steps_per_second": 9.882, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.9545631408691406, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.289, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8060710498409331, | |
| "eval_loss": 0.3684280812740326, | |
| "eval_precision": 0.7988372093023256, | |
| "eval_recall": 0.8154664484451719, | |
| "eval_runtime": 5.1019, | |
| "eval_samples_per_second": 78.206, | |
| "eval_steps_per_second": 9.8, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 5.748187065124512, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2997, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8176861216035092, | |
| "eval_loss": 0.3479882776737213, | |
| "eval_precision": 0.8193355786895284, | |
| "eval_recall": 0.8161029278050556, | |
| "eval_runtime": 5.0557, | |
| "eval_samples_per_second": 78.92, | |
| "eval_steps_per_second": 9.89, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 4.010083198547363, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2986, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.821236559139785, | |
| "eval_loss": 0.35758015513420105, | |
| "eval_precision": 0.8169406150583245, | |
| "eval_recall": 0.8261047463175123, | |
| "eval_runtime": 5.0955, | |
| "eval_samples_per_second": 78.304, | |
| "eval_steps_per_second": 9.813, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.9220337271690369, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2914, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8195005730140539, | |
| "eval_loss": 0.34965991973876953, | |
| "eval_precision": 0.8179621848739496, | |
| "eval_recall": 0.8211038370612839, | |
| "eval_runtime": 5.0617, | |
| "eval_samples_per_second": 78.827, | |
| "eval_steps_per_second": 9.878, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 1.7026562690734863, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.278, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8229427559286084, | |
| "eval_loss": 0.3539772927761078, | |
| "eval_precision": 0.8206541218637993, | |
| "eval_recall": 0.8253773413347881, | |
| "eval_runtime": 5.1199, | |
| "eval_samples_per_second": 77.931, | |
| "eval_steps_per_second": 9.766, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 5.839470863342285, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2887, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8229427559286084, | |
| "eval_loss": 0.35161107778549194, | |
| "eval_precision": 0.8206541218637993, | |
| "eval_recall": 0.8253773413347881, | |
| "eval_runtime": 5.1154, | |
| "eval_samples_per_second": 77.999, | |
| "eval_steps_per_second": 9.774, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 7.782900810241699, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2829, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8229427559286084, | |
| "eval_loss": 0.35369938611984253, | |
| "eval_precision": 0.8206541218637993, | |
| "eval_recall": 0.8253773413347881, | |
| "eval_runtime": 5.0565, | |
| "eval_samples_per_second": 78.908, | |
| "eval_steps_per_second": 9.888, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 5.059621334075928, | |
| "learning_rate": 0.0, | |
| "loss": 0.2771, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8263655462184873, | |
| "eval_loss": 0.35401326417922974, | |
| "eval_precision": 0.8233396753671443, | |
| "eval_recall": 0.8296508456082925, | |
| "eval_runtime": 5.0854, | |
| "eval_samples_per_second": 78.459, | |
| "eval_steps_per_second": 9.832, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7597037114448000.0, | |
| "train_loss": 0.35800845193081215, | |
| "train_runtime": 2113.1391, | |
| "train_samples_per_second": 34.432, | |
| "train_steps_per_second": 1.155 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7597037114448000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |