| { | |
| "best_metric": 0.8142587508579271, | |
| "best_model_checkpoint": "final_models/transformer_base_final_2/finetune/qqp/checkpoint-7600", | |
| "epoch": 4.7290640394088665, | |
| "global_step": 9600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.7154598832130432, | |
| "eval_f1": 0.5973793611535021, | |
| "eval_loss": 0.553121030330658, | |
| "eval_mcc": 0.4064984762837603, | |
| "eval_runtime": 36.7967, | |
| "eval_samples_per_second": 730.745, | |
| "eval_steps_per_second": 91.367, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.876847290640394e-05, | |
| "loss": 0.5958, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_accuracy": 0.7652571797370911, | |
| "eval_f1": 0.7188168210976479, | |
| "eval_loss": 0.4824499189853668, | |
| "eval_mcc": 0.5174020903173898, | |
| "eval_runtime": 37.2147, | |
| "eval_samples_per_second": 722.538, | |
| "eval_steps_per_second": 90.341, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.753694581280788e-05, | |
| "loss": 0.4983, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_accuracy": 0.7783480286598206, | |
| "eval_f1": 0.7443376801647221, | |
| "eval_loss": 0.4612788259983063, | |
| "eval_mcc": 0.5498175947268261, | |
| "eval_runtime": 36.7893, | |
| "eval_samples_per_second": 730.892, | |
| "eval_steps_per_second": 91.385, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.630541871921182e-05, | |
| "loss": 0.4669, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_accuracy": 0.7915132641792297, | |
| "eval_f1": 0.7595642477268828, | |
| "eval_loss": 0.445358544588089, | |
| "eval_mcc": 0.5766718430214125, | |
| "eval_runtime": 37.1991, | |
| "eval_samples_per_second": 722.841, | |
| "eval_steps_per_second": 90.379, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.507389162561577e-05, | |
| "loss": 0.4396, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.7877570986747742, | |
| "eval_f1": 0.7255854209741789, | |
| "eval_loss": 0.4518950581550598, | |
| "eval_mcc": 0.5602641007879372, | |
| "eval_runtime": 36.7715, | |
| "eval_samples_per_second": 731.245, | |
| "eval_steps_per_second": 91.429, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_accuracy": 0.808546245098114, | |
| "eval_f1": 0.7764072272411395, | |
| "eval_loss": 0.4165021777153015, | |
| "eval_mcc": 0.6094217214687762, | |
| "eval_runtime": 36.7866, | |
| "eval_samples_per_second": 730.946, | |
| "eval_steps_per_second": 91.392, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.384236453201971e-05, | |
| "loss": 0.3846, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_accuracy": 0.8110751509666443, | |
| "eval_f1": 0.7758164165931156, | |
| "eval_loss": 0.4379393756389618, | |
| "eval_mcc": 0.61257416035664, | |
| "eval_runtime": 36.9761, | |
| "eval_samples_per_second": 727.2, | |
| "eval_steps_per_second": 90.924, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 4.261083743842365e-05, | |
| "loss": 0.3731, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_accuracy": 0.8046041131019592, | |
| "eval_f1": 0.7478160698857637, | |
| "eval_loss": 0.4554973840713501, | |
| "eval_mcc": 0.5959099136582018, | |
| "eval_runtime": 37.1999, | |
| "eval_samples_per_second": 722.825, | |
| "eval_steps_per_second": 90.377, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 4.1379310344827587e-05, | |
| "loss": 0.3644, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_accuracy": 0.8124139904975891, | |
| "eval_f1": 0.763702801461632, | |
| "eval_loss": 0.41165271401405334, | |
| "eval_mcc": 0.611986827013611, | |
| "eval_runtime": 36.8489, | |
| "eval_samples_per_second": 729.709, | |
| "eval_steps_per_second": 91.237, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 4.014778325123153e-05, | |
| "loss": 0.3597, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.8246495127677917, | |
| "eval_f1": 0.7941317731301577, | |
| "eval_loss": 0.3947930634021759, | |
| "eval_mcc": 0.6416304162338577, | |
| "eval_runtime": 36.786, | |
| "eval_samples_per_second": 730.958, | |
| "eval_steps_per_second": 91.394, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_accuracy": 0.8287403583526611, | |
| "eval_f1": 0.8015171759837938, | |
| "eval_loss": 0.43710950016975403, | |
| "eval_mcc": 0.651756180893439, | |
| "eval_runtime": 37.3039, | |
| "eval_samples_per_second": 720.808, | |
| "eval_steps_per_second": 90.125, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 3.891625615763547e-05, | |
| "loss": 0.2958, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_accuracy": 0.8277362585067749, | |
| "eval_f1": 0.7931773530987676, | |
| "eval_loss": 0.44132187962532043, | |
| "eval_mcc": 0.6456804038262367, | |
| "eval_runtime": 36.7484, | |
| "eval_samples_per_second": 731.705, | |
| "eval_steps_per_second": 91.487, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 3.768472906403941e-05, | |
| "loss": 0.2907, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_accuracy": 0.8306742310523987, | |
| "eval_f1": 0.799170746769, | |
| "eval_loss": 0.40259310603141785, | |
| "eval_mcc": 0.6528157073872258, | |
| "eval_runtime": 36.7616, | |
| "eval_samples_per_second": 731.443, | |
| "eval_steps_per_second": 91.454, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 3.645320197044335e-05, | |
| "loss": 0.2897, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_accuracy": 0.8309345841407776, | |
| "eval_f1": 0.7963991400931566, | |
| "eval_loss": 0.41956043243408203, | |
| "eval_mcc": 0.652035643544486, | |
| "eval_runtime": 36.6997, | |
| "eval_samples_per_second": 732.675, | |
| "eval_steps_per_second": 91.608, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 3.522167487684729e-05, | |
| "loss": 0.2897, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.8341701030731201, | |
| "eval_f1": 0.8046440306681271, | |
| "eval_loss": 0.38309139013290405, | |
| "eval_mcc": 0.6607038531075586, | |
| "eval_runtime": 36.6961, | |
| "eval_samples_per_second": 732.748, | |
| "eval_steps_per_second": 91.617, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "eval_accuracy": 0.8317899703979492, | |
| "eval_f1": 0.795070454442481, | |
| "eval_loss": 0.4706867039203644, | |
| "eval_mcc": 0.6531092501008526, | |
| "eval_runtime": 37.0244, | |
| "eval_samples_per_second": 726.251, | |
| "eval_steps_per_second": 90.805, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 3.399014778325123e-05, | |
| "loss": 0.2317, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "eval_accuracy": 0.8349882960319519, | |
| "eval_f1": 0.8043046795748245, | |
| "eval_loss": 0.45305201411247253, | |
| "eval_mcc": 0.66167024556867, | |
| "eval_runtime": 36.9761, | |
| "eval_samples_per_second": 727.199, | |
| "eval_steps_per_second": 90.924, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 3.275862068965517e-05, | |
| "loss": 0.2273, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "eval_accuracy": 0.835248589515686, | |
| "eval_f1": 0.8068875326939843, | |
| "eval_loss": 0.4349125027656555, | |
| "eval_mcc": 0.6635037428287318, | |
| "eval_runtime": 37.2506, | |
| "eval_samples_per_second": 721.841, | |
| "eval_steps_per_second": 90.254, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 3.152709359605912e-05, | |
| "loss": 0.2257, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "eval_accuracy": 0.8389676213264465, | |
| "eval_f1": 0.8142587508579271, | |
| "eval_loss": 0.43228474259376526, | |
| "eval_mcc": 0.6733312473430861, | |
| "eval_runtime": 36.7955, | |
| "eval_samples_per_second": 730.769, | |
| "eval_steps_per_second": 91.37, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 3.0295566502463057e-05, | |
| "loss": 0.2282, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "eval_accuracy": 0.840380847454071, | |
| "eval_f1": 0.8076026537565, | |
| "eval_loss": 0.4352688193321228, | |
| "eval_mcc": 0.6714324068394563, | |
| "eval_runtime": 36.745, | |
| "eval_samples_per_second": 731.773, | |
| "eval_steps_per_second": 91.495, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_accuracy": 0.8367362022399902, | |
| "eval_f1": 0.8015370705244124, | |
| "eval_loss": 0.5048763751983643, | |
| "eval_mcc": 0.663438414164681, | |
| "eval_runtime": 36.7329, | |
| "eval_samples_per_second": 732.013, | |
| "eval_steps_per_second": 91.525, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 2.9064039408866993e-05, | |
| "loss": 0.1807, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "eval_accuracy": 0.839153528213501, | |
| "eval_f1": 0.8050132996708894, | |
| "eval_loss": 0.4986499547958374, | |
| "eval_mcc": 0.668570004117267, | |
| "eval_runtime": 36.7487, | |
| "eval_samples_per_second": 731.7, | |
| "eval_steps_per_second": 91.486, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 2.7832512315270936e-05, | |
| "loss": 0.1756, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "eval_accuracy": 0.8389304280281067, | |
| "eval_f1": 0.8094672473714312, | |
| "eval_loss": 0.4615938365459442, | |
| "eval_mcc": 0.670007984448241, | |
| "eval_runtime": 36.7794, | |
| "eval_samples_per_second": 731.089, | |
| "eval_steps_per_second": 91.41, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 2.660098522167488e-05, | |
| "loss": 0.1758, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "eval_accuracy": 0.8421287536621094, | |
| "eval_f1": 0.8122761243532481, | |
| "eval_loss": 0.5110430717468262, | |
| "eval_mcc": 0.6760658284875277, | |
| "eval_runtime": 36.9386, | |
| "eval_samples_per_second": 727.937, | |
| "eval_steps_per_second": 91.016, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "step": 9600, | |
| "total_flos": 7.577744814102528e+16, | |
| "train_loss": 0.3191047571102778, | |
| "train_runtime": 4680.9438, | |
| "train_samples_per_second": 520.19, | |
| "train_steps_per_second": 4.337 | |
| } | |
| ], | |
| "max_steps": 20300, | |
| "num_train_epochs": 10, | |
| "total_flos": 7.577744814102528e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |