| { | |
| "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": 5.094232082366943, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5608, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7218045112781954, | |
| "eval_f1": 0.6526816032372154, | |
| "eval_loss": 0.5057180523872375, | |
| "eval_precision": 0.6593383311603651, | |
| "eval_recall": 0.6481633024186215, | |
| "eval_runtime": 5.071, | |
| "eval_samples_per_second": 78.683, | |
| "eval_steps_per_second": 9.86, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.818559408187866, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5012, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7518796992481203, | |
| "eval_f1": 0.7192637077573647, | |
| "eval_loss": 0.47916004061698914, | |
| "eval_precision": 0.7116519573339108, | |
| "eval_recall": 0.7369521731223858, | |
| "eval_runtime": 5.0875, | |
| "eval_samples_per_second": 78.427, | |
| "eval_steps_per_second": 9.828, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.010869026184082, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4628, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7694235588972431, | |
| "eval_f1": 0.7320583941605839, | |
| "eval_loss": 0.4281270503997803, | |
| "eval_precision": 0.7255639097744361, | |
| "eval_recall": 0.7418621567557737, | |
| "eval_runtime": 5.0471, | |
| "eval_samples_per_second": 79.056, | |
| "eval_steps_per_second": 9.907, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.6188292503356934, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4045, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8170426065162907, | |
| "eval_f1": 0.7764637262952703, | |
| "eval_loss": 0.39506053924560547, | |
| "eval_precision": 0.7802700348432056, | |
| "eval_recall": 0.7730496453900709, | |
| "eval_runtime": 5.0784, | |
| "eval_samples_per_second": 78.569, | |
| "eval_steps_per_second": 9.846, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.2819700241088867, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3701, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7994987468671679, | |
| "eval_f1": 0.7735784814436499, | |
| "eval_loss": 0.42390376329421997, | |
| "eval_precision": 0.7633219954648527, | |
| "eval_recall": 0.7956446626659393, | |
| "eval_runtime": 5.0599, | |
| "eval_samples_per_second": 78.856, | |
| "eval_steps_per_second": 9.882, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.4958856105804443, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3362, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7974014336917563, | |
| "eval_loss": 0.3720650374889374, | |
| "eval_precision": 0.7934340756451043, | |
| "eval_recall": 0.801918530641935, | |
| "eval_runtime": 5.0782, | |
| "eval_samples_per_second": 78.572, | |
| "eval_steps_per_second": 9.846, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.6802586317062378, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3285, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8077810218978102, | |
| "eval_loss": 0.372498095035553, | |
| "eval_precision": 0.7988721804511278, | |
| "eval_recall": 0.8204673577014002, | |
| "eval_runtime": 5.076, | |
| "eval_samples_per_second": 78.605, | |
| "eval_steps_per_second": 9.85, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.754953384399414, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3061, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8109388749746158, | |
| "eval_loss": 0.3537139594554901, | |
| "eval_precision": 0.8087365591397849, | |
| "eval_recall": 0.8132842334969994, | |
| "eval_runtime": 5.0678, | |
| "eval_samples_per_second": 78.732, | |
| "eval_steps_per_second": 9.866, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 19.56822395324707, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3017, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8109388749746158, | |
| "eval_loss": 0.3503880798816681, | |
| "eval_precision": 0.8087365591397849, | |
| "eval_recall": 0.8132842334969994, | |
| "eval_runtime": 5.0696, | |
| "eval_samples_per_second": 78.704, | |
| "eval_steps_per_second": 9.863, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 7.07460880279541, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2942, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8186033824331697, | |
| "eval_loss": 0.33907514810562134, | |
| "eval_precision": 0.8186033824331697, | |
| "eval_recall": 0.8186033824331697, | |
| "eval_runtime": 5.0634, | |
| "eval_samples_per_second": 78.8, | |
| "eval_steps_per_second": 9.875, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.607100486755371, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2715, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.821236559139785, | |
| "eval_loss": 0.34563252329826355, | |
| "eval_precision": 0.8169406150583245, | |
| "eval_recall": 0.8261047463175123, | |
| "eval_runtime": 5.0549, | |
| "eval_samples_per_second": 78.934, | |
| "eval_steps_per_second": 9.891, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 10.076610565185547, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2703, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8262464479462671, | |
| "eval_loss": 0.353447824716568, | |
| "eval_precision": 0.8190008071955719, | |
| "eval_recall": 0.8353791598472449, | |
| "eval_runtime": 5.1064, | |
| "eval_samples_per_second": 78.137, | |
| "eval_steps_per_second": 9.792, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.735930323600769, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2759, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8202661535994868, | |
| "eval_loss": 0.332623153924942, | |
| "eval_precision": 0.8228070175438597, | |
| "eval_recall": 0.8178759774504456, | |
| "eval_runtime": 5.0603, | |
| "eval_samples_per_second": 78.849, | |
| "eval_steps_per_second": 9.881, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 11.147435188293457, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2705, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8289446964056049, | |
| "eval_loss": 0.33598676323890686, | |
| "eval_precision": 0.8266129032258065, | |
| "eval_recall": 0.8314238952536825, | |
| "eval_runtime": 5.0525, | |
| "eval_samples_per_second": 78.971, | |
| "eval_steps_per_second": 9.896, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.3279173374176025, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2576, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.839868600986979, | |
| "eval_loss": 0.3422936797142029, | |
| "eval_precision": 0.834029197080292, | |
| "eval_recall": 0.8467448627023095, | |
| "eval_runtime": 5.0807, | |
| "eval_samples_per_second": 78.532, | |
| "eval_steps_per_second": 9.841, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.688598155975342, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2513, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8313646980313647, | |
| "eval_loss": 0.33944278955459595, | |
| "eval_precision": 0.825091575091575, | |
| "eval_recall": 0.8389252591380251, | |
| "eval_runtime": 5.0529, | |
| "eval_samples_per_second": 78.964, | |
| "eval_steps_per_second": 9.895, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 8.046936988830566, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2481, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8281017346283209, | |
| "eval_loss": 0.32614341378211975, | |
| "eval_precision": 0.8272965800108572, | |
| "eval_recall": 0.8289234406255683, | |
| "eval_runtime": 5.0441, | |
| "eval_samples_per_second": 79.102, | |
| "eval_steps_per_second": 9.913, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.366418361663818, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2561, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8365204824303285, | |
| "eval_loss": 0.3320069909095764, | |
| "eval_precision": 0.8313636363636363, | |
| "eval_recall": 0.8424713584288053, | |
| "eval_runtime": 5.0899, | |
| "eval_samples_per_second": 78.39, | |
| "eval_steps_per_second": 9.823, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.5271737575531006, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2478, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8349466368826011, | |
| "eval_loss": 0.3268921673297882, | |
| "eval_precision": 0.8325716845878136, | |
| "eval_recall": 0.8374704491725768, | |
| "eval_runtime": 5.0459, | |
| "eval_samples_per_second": 79.074, | |
| "eval_steps_per_second": 9.909, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 10.424647331237793, | |
| "learning_rate": 0.0, | |
| "loss": 0.2451, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8357422474382676, | |
| "eval_loss": 0.32743164896965027, | |
| "eval_precision": 0.8319228265372551, | |
| "eval_recall": 0.8399709038006911, | |
| "eval_runtime": 5.0792, | |
| "eval_samples_per_second": 78.556, | |
| "eval_steps_per_second": 9.844, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7609911792720000.0, | |
| "train_loss": 0.32301358082255377, | |
| "train_runtime": 1951.399, | |
| "train_samples_per_second": 37.286, | |
| "train_steps_per_second": 1.25 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7609911792720000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |