| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.668501853942871, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.2459, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8352737181075056, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.7375997304916382, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 0.9353, | |
| "eval_samples_per_second": 199.938, | |
| "eval_steps_per_second": 3.208, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.2960220575332642, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.7125, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8362645528858063, | |
| "eval_f1": 0.005681818181818181, | |
| "eval_loss": 0.6395161151885986, | |
| "eval_precision": 0.16666666666666666, | |
| "eval_recall": 0.002890173410404624, | |
| "eval_runtime": 0.9886, | |
| "eval_samples_per_second": 189.166, | |
| "eval_steps_per_second": 3.035, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.4184564352035522, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.6362, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.839980183304434, | |
| "eval_f1": 0.02168021680216802, | |
| "eval_loss": 0.5518138408660889, | |
| "eval_precision": 0.17391304347826086, | |
| "eval_recall": 0.011560693641618497, | |
| "eval_runtime": 1.0186, | |
| "eval_samples_per_second": 183.577, | |
| "eval_steps_per_second": 2.945, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.0016016960144043, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5564, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8578152093138469, | |
| "eval_f1": 0.11389521640091117, | |
| "eval_loss": 0.46721315383911133, | |
| "eval_precision": 0.26881720430107525, | |
| "eval_recall": 0.07225433526011561, | |
| "eval_runtime": 0.9821, | |
| "eval_samples_per_second": 190.403, | |
| "eval_steps_per_second": 3.055, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.4191057682037354, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4714, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8880356700520188, | |
| "eval_f1": 0.3236363636363636, | |
| "eval_loss": 0.3911801874637604, | |
| "eval_precision": 0.4362745098039216, | |
| "eval_recall": 0.25722543352601157, | |
| "eval_runtime": 0.9404, | |
| "eval_samples_per_second": 198.861, | |
| "eval_steps_per_second": 3.19, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.1648097038269043, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3978, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9135496655932623, | |
| "eval_f1": 0.5105740181268881, | |
| "eval_loss": 0.324018269777298, | |
| "eval_precision": 0.5348101265822784, | |
| "eval_recall": 0.4884393063583815, | |
| "eval_runtime": 0.9314, | |
| "eval_samples_per_second": 200.778, | |
| "eval_steps_per_second": 3.221, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.7792711853981018, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3365, | |
| "step": 742 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.924201139459995, | |
| "eval_f1": 0.6122448979591837, | |
| "eval_loss": 0.28386977314949036, | |
| "eval_precision": 0.5784061696658098, | |
| "eval_recall": 0.6502890173410405, | |
| "eval_runtime": 0.9336, | |
| "eval_samples_per_second": 200.302, | |
| "eval_steps_per_second": 3.213, | |
| "step": 742 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.4722111225128174, | |
| "learning_rate": 3e-05, | |
| "loss": 0.294, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9318801089918256, | |
| "eval_f1": 0.6657789613848202, | |
| "eval_loss": 0.2507442831993103, | |
| "eval_precision": 0.6172839506172839, | |
| "eval_recall": 0.7225433526011561, | |
| "eval_runtime": 0.9589, | |
| "eval_samples_per_second": 195.017, | |
| "eval_steps_per_second": 3.129, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.440127968788147, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2677, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9355957394104533, | |
| "eval_f1": 0.6973684210526315, | |
| "eval_loss": 0.23195113241672516, | |
| "eval_precision": 0.6400966183574879, | |
| "eval_recall": 0.7658959537572254, | |
| "eval_runtime": 0.9704, | |
| "eval_samples_per_second": 192.695, | |
| "eval_steps_per_second": 3.091, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.053550124168396, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2457, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.939311369829081, | |
| "eval_f1": 0.716180371352785, | |
| "eval_loss": 0.2109125703573227, | |
| "eval_precision": 0.6617647058823529, | |
| "eval_recall": 0.7803468208092486, | |
| "eval_runtime": 0.9655, | |
| "eval_samples_per_second": 193.689, | |
| "eval_steps_per_second": 3.107, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.8145582675933838, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2339, | |
| "step": 1166 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9405499133019569, | |
| "eval_f1": 0.7239101717305151, | |
| "eval_loss": 0.20224529504776, | |
| "eval_precision": 0.6666666666666666, | |
| "eval_recall": 0.791907514450867, | |
| "eval_runtime": 0.9648, | |
| "eval_samples_per_second": 193.823, | |
| "eval_steps_per_second": 3.109, | |
| "step": 1166 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.759846806526184, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2215, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9425315828585583, | |
| "eval_f1": 0.7450980392156863, | |
| "eval_loss": 0.1986953467130661, | |
| "eval_precision": 0.6801909307875895, | |
| "eval_recall": 0.8236994219653179, | |
| "eval_runtime": 1.023, | |
| "eval_samples_per_second": 182.798, | |
| "eval_steps_per_second": 2.933, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.7489528656005859, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2125, | |
| "step": 1378 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9432747089422838, | |
| "eval_f1": 0.7408376963350786, | |
| "eval_loss": 0.18985998630523682, | |
| "eval_precision": 0.6770334928229665, | |
| "eval_recall": 0.8179190751445087, | |
| "eval_runtime": 0.9458, | |
| "eval_samples_per_second": 197.715, | |
| "eval_steps_per_second": 3.172, | |
| "step": 1378 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.942519187927246, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2085, | |
| "step": 1484 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9437701263314342, | |
| "eval_f1": 0.7463863337713535, | |
| "eval_loss": 0.1853644996881485, | |
| "eval_precision": 0.6843373493975904, | |
| "eval_recall": 0.8208092485549133, | |
| "eval_runtime": 0.9518, | |
| "eval_samples_per_second": 196.479, | |
| "eval_steps_per_second": 3.152, | |
| "step": 1484 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.6467780470848083, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2002, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9459995045826108, | |
| "eval_f1": 0.7519788918205805, | |
| "eval_loss": 0.17970043420791626, | |
| "eval_precision": 0.691747572815534, | |
| "eval_recall": 0.8236994219653179, | |
| "eval_runtime": 0.9531, | |
| "eval_samples_per_second": 196.206, | |
| "eval_steps_per_second": 3.148, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.3061821460723877, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2, | |
| "step": 1696 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9452563784988853, | |
| "eval_f1": 0.7490144546649146, | |
| "eval_loss": 0.1779414862394333, | |
| "eval_precision": 0.6867469879518072, | |
| "eval_recall": 0.8236994219653179, | |
| "eval_runtime": 0.9553, | |
| "eval_samples_per_second": 195.76, | |
| "eval_steps_per_second": 3.141, | |
| "step": 1696 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.871351420879364, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1929, | |
| "step": 1802 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9450086698043101, | |
| "eval_f1": 0.7486910994764396, | |
| "eval_loss": 0.177422896027565, | |
| "eval_precision": 0.6842105263157895, | |
| "eval_recall": 0.8265895953757225, | |
| "eval_runtime": 1.0338, | |
| "eval_samples_per_second": 180.886, | |
| "eval_steps_per_second": 2.902, | |
| "step": 1802 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 1.4120484590530396, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1932, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9457517958880357, | |
| "eval_f1": 0.7506561679790026, | |
| "eval_loss": 0.1760500967502594, | |
| "eval_precision": 0.6875, | |
| "eval_recall": 0.8265895953757225, | |
| "eval_runtime": 0.9548, | |
| "eval_samples_per_second": 195.862, | |
| "eval_steps_per_second": 3.142, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.9899640679359436, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1916, | |
| "step": 2014 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9464949219717612, | |
| "eval_f1": 0.7516425755584757, | |
| "eval_loss": 0.17468006908893585, | |
| "eval_precision": 0.689156626506024, | |
| "eval_recall": 0.8265895953757225, | |
| "eval_runtime": 0.9571, | |
| "eval_samples_per_second": 195.387, | |
| "eval_steps_per_second": 3.135, | |
| "step": 2014 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.6105211973190308, | |
| "learning_rate": 0.0, | |
| "loss": 0.1887, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9464949219717612, | |
| "eval_f1": 0.7516425755584757, | |
| "eval_loss": 0.17423608899116516, | |
| "eval_precision": 0.689156626506024, | |
| "eval_recall": 0.8265895953757225, | |
| "eval_runtime": 0.9545, | |
| "eval_samples_per_second": 195.923, | |
| "eval_steps_per_second": 3.143, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2120, | |
| "total_flos": 910226579721144.0, | |
| "train_loss": 0.36036560220538444, | |
| "train_runtime": 247.3916, | |
| "train_samples_per_second": 136.383, | |
| "train_steps_per_second": 8.569 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 910226579721144.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |