| { | |
| "best_metric": 0.8454404945904173, | |
| "best_model_checkpoint": "cvt-13-384-in22k-FV-finetuned-memes/checkpoint-180", | |
| "epoch": 19.987654320987655, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3e-05, | |
| "loss": 1.6621, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 6e-05, | |
| "loss": 1.4066, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.5123647604327666, | |
| "eval_f1": 0.43706156197054624, | |
| "eval_loss": 1.2430025339126587, | |
| "eval_precision": 0.5141456021674624, | |
| "eval_recall": 0.5123647604327666, | |
| "eval_runtime": 12.004, | |
| "eval_samples_per_second": 107.798, | |
| "eval_steps_per_second": 1.749, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 9e-05, | |
| "loss": 1.2862, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00012, | |
| "loss": 1.0813, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.6893353941267388, | |
| "eval_f1": 0.6615579665180347, | |
| "eval_loss": 0.8244319558143616, | |
| "eval_precision": 0.6834239635195151, | |
| "eval_recall": 0.6893353941267388, | |
| "eval_runtime": 11.1665, | |
| "eval_samples_per_second": 115.882, | |
| "eval_steps_per_second": 1.881, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00011666666666666667, | |
| "loss": 0.9199, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.00011333333333333333, | |
| "loss": 0.8392, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.7612055641421948, | |
| "eval_f1": 0.7569545008442342, | |
| "eval_loss": 0.6334171891212463, | |
| "eval_precision": 0.767024313347433, | |
| "eval_recall": 0.7612055641421948, | |
| "eval_runtime": 10.5864, | |
| "eval_samples_per_second": 122.232, | |
| "eval_steps_per_second": 1.984, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 0.00011, | |
| "loss": 0.7028, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 0.7065, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "eval_accuracy": 0.776661514683153, | |
| "eval_f1": 0.7671672909038773, | |
| "eval_loss": 0.581936240196228, | |
| "eval_precision": 0.7799046353657799, | |
| "eval_recall": 0.776661514683153, | |
| "eval_runtime": 10.9056, | |
| "eval_samples_per_second": 118.654, | |
| "eval_steps_per_second": 1.926, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 0.00010333333333333334, | |
| "loss": 0.6294, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5751, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.8176197836166924, | |
| "eval_f1": 0.8130327497943682, | |
| "eval_loss": 0.5365290641784668, | |
| "eval_precision": 0.8215999410934138, | |
| "eval_recall": 0.8176197836166924, | |
| "eval_runtime": 10.6877, | |
| "eval_samples_per_second": 121.074, | |
| "eval_steps_per_second": 1.965, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 9.666666666666667e-05, | |
| "loss": 0.5597, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 0.4896, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.8307573415765069, | |
| "eval_f1": 0.8264996195118711, | |
| "eval_loss": 0.49427157640457153, | |
| "eval_precision": 0.8257369667433976, | |
| "eval_recall": 0.8307573415765069, | |
| "eval_runtime": 11.8177, | |
| "eval_samples_per_second": 109.496, | |
| "eval_steps_per_second": 1.777, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 9e-05, | |
| "loss": 0.4976, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 8.666666666666667e-05, | |
| "loss": 0.4487, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "eval_accuracy": 0.8106646058732612, | |
| "eval_f1": 0.8053637875370054, | |
| "eval_loss": 0.5398644208908081, | |
| "eval_precision": 0.806886049971449, | |
| "eval_recall": 0.8106646058732612, | |
| "eval_runtime": 11.1665, | |
| "eval_samples_per_second": 115.882, | |
| "eval_steps_per_second": 1.881, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.4755, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 7.999999999999999e-05, | |
| "loss": 0.4349, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "eval_accuracy": 0.8299845440494591, | |
| "eval_f1": 0.8273441546369988, | |
| "eval_loss": 0.48919445276260376, | |
| "eval_precision": 0.8285069417879617, | |
| "eval_recall": 0.8299845440494591, | |
| "eval_runtime": 10.9335, | |
| "eval_samples_per_second": 118.351, | |
| "eval_steps_per_second": 1.921, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 7.666666666666667e-05, | |
| "loss": 0.4266, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "learning_rate": 7.333333333333334e-05, | |
| "loss": 0.43, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.8454404945904173, | |
| "eval_f1": 0.8425612552932723, | |
| "eval_loss": 0.4984479546546936, | |
| "eval_precision": 0.8464868510362556, | |
| "eval_recall": 0.8454404945904173, | |
| "eval_runtime": 10.7938, | |
| "eval_samples_per_second": 119.884, | |
| "eval_steps_per_second": 1.946, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "learning_rate": 7.000000000000001e-05, | |
| "loss": 0.4143, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.4372, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "eval_accuracy": 0.8191653786707882, | |
| "eval_f1": 0.8157445071524905, | |
| "eval_loss": 0.5572782158851624, | |
| "eval_precision": 0.822149098583948, | |
| "eval_recall": 0.8191653786707882, | |
| "eval_runtime": 10.9137, | |
| "eval_samples_per_second": 118.567, | |
| "eval_steps_per_second": 1.924, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 6.333333333333333e-05, | |
| "loss": 0.3865, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "learning_rate": 6e-05, | |
| "loss": 0.3994, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_accuracy": 0.8299845440494591, | |
| "eval_f1": 0.82808018136039, | |
| "eval_loss": 0.5158317685127258, | |
| "eval_precision": 0.8283803654792194, | |
| "eval_recall": 0.8299845440494591, | |
| "eval_runtime": 11.8018, | |
| "eval_samples_per_second": 109.644, | |
| "eval_steps_per_second": 1.779, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 5.6666666666666664e-05, | |
| "loss": 0.3822, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.3883, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "eval_accuracy": 0.8353941267387944, | |
| "eval_f1": 0.8314264954771738, | |
| "eval_loss": 0.5494962930679321, | |
| "eval_precision": 0.8317149547063074, | |
| "eval_recall": 0.8353941267387944, | |
| "eval_runtime": 11.0963, | |
| "eval_samples_per_second": 116.616, | |
| "eval_steps_per_second": 1.893, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3857, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.406, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.8284389489953632, | |
| "eval_f1": 0.8245717709756429, | |
| "eval_loss": 0.5297971367835999, | |
| "eval_precision": 0.8285366488073825, | |
| "eval_recall": 0.8284389489953632, | |
| "eval_runtime": 11.0958, | |
| "eval_samples_per_second": 116.621, | |
| "eval_steps_per_second": 1.893, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 13.49, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.3594, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 3.9999999999999996e-05, | |
| "loss": 0.3355, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "eval_accuracy": 0.839258114374034, | |
| "eval_f1": 0.8356831660968493, | |
| "eval_loss": 0.5400787591934204, | |
| "eval_precision": 0.8345744148504253, | |
| "eval_recall": 0.839258114374034, | |
| "eval_runtime": 10.7779, | |
| "eval_samples_per_second": 120.06, | |
| "eval_steps_per_second": 1.948, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 3.666666666666667e-05, | |
| "loss": 0.3706, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.395, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "eval_accuracy": 0.8307573415765069, | |
| "eval_f1": 0.8260821652476246, | |
| "eval_loss": 0.5915156602859497, | |
| "eval_precision": 0.8278203282316018, | |
| "eval_recall": 0.8307573415765069, | |
| "eval_runtime": 10.988, | |
| "eval_samples_per_second": 117.765, | |
| "eval_steps_per_second": 1.911, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3548, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.3612, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "eval_accuracy": 0.8408037094281299, | |
| "eval_f1": 0.8368203137682052, | |
| "eval_loss": 0.5851544737815857, | |
| "eval_precision": 0.8377711910324739, | |
| "eval_recall": 0.8408037094281299, | |
| "eval_runtime": 11.7394, | |
| "eval_samples_per_second": 110.227, | |
| "eval_steps_per_second": 1.789, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 16.49, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.3594, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "learning_rate": 1.9999999999999998e-05, | |
| "loss": 0.3765, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.8384853168469861, | |
| "eval_f1": 0.8356408095957625, | |
| "eval_loss": 0.5508687496185303, | |
| "eval_precision": 0.8351154357408708, | |
| "eval_recall": 0.8384853168469861, | |
| "eval_runtime": 10.7561, | |
| "eval_samples_per_second": 120.304, | |
| "eval_steps_per_second": 1.952, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.3896, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.3688, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "eval_accuracy": 0.8415765069551777, | |
| "eval_f1": 0.8386603858270167, | |
| "eval_loss": 0.5668028593063354, | |
| "eval_precision": 0.8398264475825991, | |
| "eval_recall": 0.8415765069551777, | |
| "eval_runtime": 10.5834, | |
| "eval_samples_per_second": 122.266, | |
| "eval_steps_per_second": 1.984, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 18.49, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 0.3705, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.3503, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_accuracy": 0.839258114374034, | |
| "eval_f1": 0.8364792101800207, | |
| "eval_loss": 0.5626189708709717, | |
| "eval_precision": 0.8370661747552312, | |
| "eval_recall": 0.839258114374034, | |
| "eval_runtime": 10.7269, | |
| "eval_samples_per_second": 120.631, | |
| "eval_steps_per_second": 1.958, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 19.49, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.3613, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "learning_rate": 0.0, | |
| "loss": 0.3611, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "eval_accuracy": 0.8346213292117465, | |
| "eval_f1": 0.8322067261008879, | |
| "eval_loss": 0.5595440864562988, | |
| "eval_precision": 0.8326806465391725, | |
| "eval_recall": 0.8346213292117465, | |
| "eval_runtime": 10.6902, | |
| "eval_samples_per_second": 121.045, | |
| "eval_steps_per_second": 1.964, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "step": 400, | |
| "total_flos": 1.832202236159447e+18, | |
| "train_loss": 0.5471319568157196, | |
| "train_runtime": 1318.2665, | |
| "train_samples_per_second": 78.497, | |
| "train_steps_per_second": 0.303 | |
| } | |
| ], | |
| "max_steps": 400, | |
| "num_train_epochs": 20, | |
| "total_flos": 1.832202236159447e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |