| { | |
| "best_metric": 0.9714, | |
| "best_model_checkpoint": "models/distilcamembert-allocine\\checkpoint-5500", | |
| "epoch": 3.0, | |
| "global_step": 7500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.0000000000000001e-07, | |
| "loss": 0.7377, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.5782, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1837, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.1665, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1504, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.95545, | |
| "eval_f1": 0.9541784520442272, | |
| "eval_loss": 0.1289823055267334, | |
| "eval_precision": 0.9614467820499534, | |
| "eval_recall": 0.9470191915067374, | |
| "eval_runtime": 51.8793, | |
| "eval_samples_per_second": 385.51, | |
| "eval_steps_per_second": 24.094, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.910714285714286e-05, | |
| "loss": 0.1469, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.8214285714285716e-05, | |
| "loss": 0.1363, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.732142857142857e-05, | |
| "loss": 0.1332, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.642857142857143e-05, | |
| "loss": 0.1334, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.96235, | |
| "eval_f1": 0.9618984971917219, | |
| "eval_loss": 0.10494959354400635, | |
| "eval_precision": 0.9536470352162135, | |
| "eval_recall": 0.9702939975500204, | |
| "eval_runtime": 52.9168, | |
| "eval_samples_per_second": 377.951, | |
| "eval_steps_per_second": 23.622, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.5535714285714286e-05, | |
| "loss": 0.1286, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.1322, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.375e-05, | |
| "loss": 0.1214, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 0.1158, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.963, | |
| "eval_f1": 0.962742926190716, | |
| "eval_loss": 0.10521914064884186, | |
| "eval_precision": 0.9498311146433539, | |
| "eval_recall": 0.9760106165781952, | |
| "eval_runtime": 52.1412, | |
| "eval_samples_per_second": 383.574, | |
| "eval_steps_per_second": 23.973, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.196428571428572e-05, | |
| "loss": 0.1146, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.107142857142857e-05, | |
| "loss": 0.1154, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.017857142857143e-05, | |
| "loss": 0.1246, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 3.928571428571429e-05, | |
| "loss": 0.1153, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.9661, | |
| "eval_f1": 0.9652770664754685, | |
| "eval_loss": 0.09492386132478714, | |
| "eval_precision": 0.9685508735868448, | |
| "eval_recall": 0.9620253164556962, | |
| "eval_runtime": 53.0859, | |
| "eval_samples_per_second": 376.748, | |
| "eval_steps_per_second": 23.547, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.839285714285715e-05, | |
| "loss": 0.1064, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.112, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.6607142857142853e-05, | |
| "loss": 0.1062, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.1053, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9666, | |
| "eval_f1": 0.9663374319693611, | |
| "eval_loss": 0.09356806427240372, | |
| "eval_precision": 0.95421974522293, | |
| "eval_recall": 0.9787668436096366, | |
| "eval_runtime": 52.3106, | |
| "eval_samples_per_second": 382.331, | |
| "eval_steps_per_second": 23.896, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 3.4821428571428574e-05, | |
| "loss": 0.0797, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3.392857142857143e-05, | |
| "loss": 0.0684, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.303571428571429e-05, | |
| "loss": 0.0745, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.2142857142857144e-05, | |
| "loss": 0.0755, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.97, | |
| "eval_f1": 0.9695400548279013, | |
| "eval_loss": 0.09874136000871658, | |
| "eval_precision": 0.964350636235104, | |
| "eval_recall": 0.9747856267864434, | |
| "eval_runtime": 52.4682, | |
| "eval_samples_per_second": 381.183, | |
| "eval_steps_per_second": 23.824, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.0771, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.0357142857142857e-05, | |
| "loss": 0.0733, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.9464285714285718e-05, | |
| "loss": 0.069, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 0.0716, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_accuracy": 0.9688, | |
| "eval_f1": 0.9684370257966616, | |
| "eval_loss": 0.10781414806842804, | |
| "eval_precision": 0.9597954682173652, | |
| "eval_recall": 0.9772356063699469, | |
| "eval_runtime": 52.2793, | |
| "eval_samples_per_second": 382.561, | |
| "eval_steps_per_second": 23.91, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 2.767857142857143e-05, | |
| "loss": 0.0781, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 0.0725, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 2.5892857142857148e-05, | |
| "loss": 0.0755, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0688, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.9673, | |
| "eval_f1": 0.967032967032967, | |
| "eval_loss": 0.10506118088960648, | |
| "eval_precision": 0.9551882095200159, | |
| "eval_recall": 0.9791751735402205, | |
| "eval_runtime": 52.2828, | |
| "eval_samples_per_second": 382.535, | |
| "eval_steps_per_second": 23.908, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 2.4107142857142858e-05, | |
| "loss": 0.07, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 2.3214285714285715e-05, | |
| "loss": 0.0735, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 2.2321428571428575e-05, | |
| "loss": 0.0748, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 0.0691, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.97095, | |
| "eval_f1": 0.9703949044585987, | |
| "eval_loss": 0.09402387589216232, | |
| "eval_precision": 0.9687658968358938, | |
| "eval_recall": 0.972029399755002, | |
| "eval_runtime": 52.5632, | |
| "eval_samples_per_second": 380.494, | |
| "eval_steps_per_second": 23.781, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 2.0535714285714285e-05, | |
| "loss": 0.0655, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.9642857142857145e-05, | |
| "loss": 0.0714, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.0783, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.0733, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.96855, | |
| "eval_f1": 0.9683171309122047, | |
| "eval_loss": 0.10380826145410538, | |
| "eval_precision": 0.9557522123893806, | |
| "eval_recall": 0.9812168231931401, | |
| "eval_runtime": 52.269, | |
| "eval_samples_per_second": 382.636, | |
| "eval_steps_per_second": 23.915, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.6964285714285715e-05, | |
| "loss": 0.0461, | |
| "step": 5125 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 1.6071428571428572e-05, | |
| "loss": 0.0451, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 1.5178571428571429e-05, | |
| "loss": 0.0502, | |
| "step": 5375 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.0476, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_accuracy": 0.9714, | |
| "eval_f1": 0.9709909727152854, | |
| "eval_loss": 0.10657692700624466, | |
| "eval_precision": 0.9648256399919372, | |
| "eval_recall": 0.9772356063699469, | |
| "eval_runtime": 51.6788, | |
| "eval_samples_per_second": 387.006, | |
| "eval_steps_per_second": 24.188, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.3392857142857144e-05, | |
| "loss": 0.0522, | |
| "step": 5625 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0476, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 1.1607142857142857e-05, | |
| "loss": 0.0404, | |
| "step": 5875 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.0714285714285714e-05, | |
| "loss": 0.047, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_accuracy": 0.96895, | |
| "eval_f1": 0.9686316108501288, | |
| "eval_loss": 0.10983184725046158, | |
| "eval_precision": 0.9587041295870413, | |
| "eval_recall": 0.9787668436096366, | |
| "eval_runtime": 52.0939, | |
| "eval_samples_per_second": 383.922, | |
| "eval_steps_per_second": 23.995, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 9.821428571428573e-06, | |
| "loss": 0.0426, | |
| "step": 6125 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 0.0396, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 8.035714285714286e-06, | |
| "loss": 0.0481, | |
| "step": 6375 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 0.0431, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_accuracy": 0.9711, | |
| "eval_f1": 0.9706211243265224, | |
| "eval_loss": 0.11103978008031845, | |
| "eval_precision": 0.9665924276169265, | |
| "eval_recall": 0.9746835443037974, | |
| "eval_runtime": 54.7325, | |
| "eval_samples_per_second": 365.413, | |
| "eval_steps_per_second": 22.838, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.0396, | |
| "step": 6625 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 5.357142857142857e-06, | |
| "loss": 0.0446, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 4.464285714285715e-06, | |
| "loss": 0.0362, | |
| "step": 6875 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 3.5714285714285714e-06, | |
| "loss": 0.0464, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_accuracy": 0.9697, | |
| "eval_f1": 0.9693970306029694, | |
| "eval_loss": 0.11486733704805374, | |
| "eval_precision": 0.9592244653208075, | |
| "eval_recall": 0.9797876684360963, | |
| "eval_runtime": 52.2973, | |
| "eval_samples_per_second": 382.429, | |
| "eval_steps_per_second": 23.902, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 2.6785714285714285e-06, | |
| "loss": 0.0431, | |
| "step": 7125 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.7857142857142857e-06, | |
| "loss": 0.0343, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 8.928571428571428e-07, | |
| "loss": 0.0467, | |
| "step": 7375 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0342, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9703, | |
| "eval_f1": 0.9699270959902794, | |
| "eval_loss": 0.11218974739313126, | |
| "eval_precision": 0.9621333869023704, | |
| "eval_recall": 0.9778481012658228, | |
| "eval_runtime": 52.6026, | |
| "eval_samples_per_second": 380.209, | |
| "eval_steps_per_second": 23.763, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 7500, | |
| "total_flos": 4.553211650587354e+16, | |
| "train_loss": 0.08937127710183461, | |
| "train_runtime": 4426.0374, | |
| "train_samples_per_second": 108.449, | |
| "train_steps_per_second": 1.695 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9714, | |
| "eval_f1": 0.9709909727152854, | |
| "eval_loss": 0.10657692700624466, | |
| "eval_precision": 0.9648256399919372, | |
| "eval_recall": 0.9772356063699469, | |
| "eval_runtime": 52.9584, | |
| "eval_samples_per_second": 377.655, | |
| "eval_steps_per_second": 23.603, | |
| "step": 7500 | |
| } | |
| ], | |
| "max_steps": 7500, | |
| "num_train_epochs": 3, | |
| "total_flos": 4.553211650587354e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |