| { | |
| "best_metric": 0.1436834264468159, | |
| "best_model_checkpoint": "cola-pixel-handwritten-mean-vatrpp-256-64-4-5e-5-15000-42/checkpoint-1600", | |
| "epoch": 78.77611940298507, | |
| "global_step": 2600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6426, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_loss": 0.6255418658256531, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 8.2752, | |
| "eval_samples_per_second": 126.039, | |
| "eval_steps_per_second": 15.83, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6176, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "eval_loss": 0.630817711353302, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 8.1785, | |
| "eval_samples_per_second": 127.53, | |
| "eval_steps_per_second": 16.018, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 4.9662162162162164e-05, | |
| "loss": 0.6183, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_loss": 0.6186701059341431, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 7.8187, | |
| "eval_samples_per_second": 133.398, | |
| "eval_steps_per_second": 16.755, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 12.12, | |
| "learning_rate": 4.9324324324324325e-05, | |
| "loss": 0.6162, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 12.12, | |
| "eval_loss": 0.6158396005630493, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 7.9658, | |
| "eval_samples_per_second": 130.935, | |
| "eval_steps_per_second": 16.445, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 15.15, | |
| "learning_rate": 4.8986486486486486e-05, | |
| "loss": 0.614, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 15.15, | |
| "eval_loss": 0.6250273585319519, | |
| "eval_matthews_correlation": -0.02929206145132745, | |
| "eval_runtime": 7.9393, | |
| "eval_samples_per_second": 131.373, | |
| "eval_steps_per_second": 16.5, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 4.8648648648648654e-05, | |
| "loss": 0.6096, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "eval_loss": 0.618541419506073, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 7.9167, | |
| "eval_samples_per_second": 131.746, | |
| "eval_steps_per_second": 16.547, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 21.21, | |
| "learning_rate": 4.8310810810810816e-05, | |
| "loss": 0.6055, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 21.21, | |
| "eval_loss": 0.6223562359809875, | |
| "eval_matthews_correlation": 0.017470726784935973, | |
| "eval_runtime": 7.909, | |
| "eval_samples_per_second": 131.875, | |
| "eval_steps_per_second": 16.563, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 24.24, | |
| "learning_rate": 4.7979729729729736e-05, | |
| "loss": 0.6001, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 24.24, | |
| "eval_loss": 0.6550844311714172, | |
| "eval_matthews_correlation": 0.13008617806157513, | |
| "eval_runtime": 8.0719, | |
| "eval_samples_per_second": 129.214, | |
| "eval_steps_per_second": 16.229, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 4.764527027027027e-05, | |
| "loss": 0.5909, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "eval_loss": 0.6533612012863159, | |
| "eval_matthews_correlation": 0.056561954805504064, | |
| "eval_runtime": 8.0431, | |
| "eval_samples_per_second": 129.676, | |
| "eval_steps_per_second": 16.287, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 30.3, | |
| "learning_rate": 4.730743243243244e-05, | |
| "loss": 0.5726, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 30.3, | |
| "eval_loss": 0.6678630709648132, | |
| "eval_matthews_correlation": 0.10285544103286985, | |
| "eval_runtime": 7.9909, | |
| "eval_samples_per_second": 130.524, | |
| "eval_steps_per_second": 16.394, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 4.697297297297297e-05, | |
| "loss": 0.5524, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "eval_loss": 0.6901304721832275, | |
| "eval_matthews_correlation": 0.06313184843830015, | |
| "eval_runtime": 8.3103, | |
| "eval_samples_per_second": 125.507, | |
| "eval_steps_per_second": 15.764, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "learning_rate": 4.663513513513514e-05, | |
| "loss": 0.5167, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "eval_loss": 0.7026833891868591, | |
| "eval_matthews_correlation": 0.09479027897712053, | |
| "eval_runtime": 8.0473, | |
| "eval_samples_per_second": 129.608, | |
| "eval_steps_per_second": 16.279, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 39.39, | |
| "learning_rate": 4.6297297297297295e-05, | |
| "loss": 0.4779, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 39.39, | |
| "eval_loss": 0.7578131556510925, | |
| "eval_matthews_correlation": 0.10121883963858187, | |
| "eval_runtime": 8.2334, | |
| "eval_samples_per_second": 126.679, | |
| "eval_steps_per_second": 15.911, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 42.42, | |
| "learning_rate": 4.5959459459459463e-05, | |
| "loss": 0.4271, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 42.42, | |
| "eval_loss": 0.8021395802497864, | |
| "eval_matthews_correlation": 0.11075235134282446, | |
| "eval_runtime": 8.0267, | |
| "eval_samples_per_second": 129.941, | |
| "eval_steps_per_second": 16.32, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 45.45, | |
| "learning_rate": 4.5621621621621625e-05, | |
| "loss": 0.3888, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 45.45, | |
| "eval_loss": 0.8813392519950867, | |
| "eval_matthews_correlation": 0.10250522978751038, | |
| "eval_runtime": 7.9207, | |
| "eval_samples_per_second": 131.68, | |
| "eval_steps_per_second": 16.539, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 48.48, | |
| "learning_rate": 4.5283783783783786e-05, | |
| "loss": 0.3428, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 48.48, | |
| "eval_loss": 0.9361783862113953, | |
| "eval_matthews_correlation": 0.1436834264468159, | |
| "eval_runtime": 7.8976, | |
| "eval_samples_per_second": 132.066, | |
| "eval_steps_per_second": 16.587, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 51.51, | |
| "learning_rate": 4.494594594594595e-05, | |
| "loss": 0.2977, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 51.51, | |
| "eval_loss": 1.078627347946167, | |
| "eval_matthews_correlation": 0.11181658143300324, | |
| "eval_runtime": 7.9966, | |
| "eval_samples_per_second": 130.43, | |
| "eval_steps_per_second": 16.382, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 54.54, | |
| "learning_rate": 4.460810810810811e-05, | |
| "loss": 0.2642, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 54.54, | |
| "eval_loss": 1.0609544515609741, | |
| "eval_matthews_correlation": 0.09013603883941315, | |
| "eval_runtime": 7.8862, | |
| "eval_samples_per_second": 132.257, | |
| "eval_steps_per_second": 16.611, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 57.57, | |
| "learning_rate": 4.427027027027027e-05, | |
| "loss": 0.2272, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 57.57, | |
| "eval_loss": 1.183494210243225, | |
| "eval_matthews_correlation": 0.11545854045964393, | |
| "eval_runtime": 7.9453, | |
| "eval_samples_per_second": 131.273, | |
| "eval_steps_per_second": 16.488, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 60.6, | |
| "learning_rate": 4.393243243243244e-05, | |
| "loss": 0.1915, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 60.6, | |
| "eval_loss": 1.2531063556671143, | |
| "eval_matthews_correlation": 0.12244528836896967, | |
| "eval_runtime": 8.0199, | |
| "eval_samples_per_second": 130.051, | |
| "eval_steps_per_second": 16.334, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 63.63, | |
| "learning_rate": 4.359459459459459e-05, | |
| "loss": 0.1691, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 63.63, | |
| "eval_loss": 1.3903430700302124, | |
| "eval_matthews_correlation": 0.07541181195571064, | |
| "eval_runtime": 7.9376, | |
| "eval_samples_per_second": 131.4, | |
| "eval_steps_per_second": 16.504, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 66.66, | |
| "learning_rate": 4.325675675675676e-05, | |
| "loss": 0.1491, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 66.66, | |
| "eval_loss": 1.4947072267532349, | |
| "eval_matthews_correlation": 0.06742580984707468, | |
| "eval_runtime": 7.9502, | |
| "eval_samples_per_second": 131.191, | |
| "eval_steps_per_second": 16.477, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 69.69, | |
| "learning_rate": 4.291891891891892e-05, | |
| "loss": 0.1339, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 69.69, | |
| "eval_loss": 1.5433533191680908, | |
| "eval_matthews_correlation": 0.0736455413240434, | |
| "eval_runtime": 8.2493, | |
| "eval_samples_per_second": 126.434, | |
| "eval_steps_per_second": 15.88, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 72.72, | |
| "learning_rate": 4.258108108108108e-05, | |
| "loss": 0.1164, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 72.72, | |
| "eval_loss": 1.5793086290359497, | |
| "eval_matthews_correlation": 0.11645872984022461, | |
| "eval_runtime": 8.2848, | |
| "eval_samples_per_second": 125.893, | |
| "eval_steps_per_second": 15.812, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 75.75, | |
| "learning_rate": 4.2243243243243244e-05, | |
| "loss": 0.1078, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 75.75, | |
| "eval_loss": 1.5938163995742798, | |
| "eval_matthews_correlation": 0.09946100084630931, | |
| "eval_runtime": 8.3263, | |
| "eval_samples_per_second": 125.266, | |
| "eval_steps_per_second": 15.733, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 78.78, | |
| "learning_rate": 4.1905405405405406e-05, | |
| "loss": 0.0974, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 78.78, | |
| "eval_loss": 1.700919508934021, | |
| "eval_matthews_correlation": 0.07568068132313144, | |
| "eval_runtime": 8.4149, | |
| "eval_samples_per_second": 123.947, | |
| "eval_steps_per_second": 15.568, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 78.78, | |
| "step": 2600, | |
| "total_flos": 1.7353653717551284e+19, | |
| "train_loss": 0.40566940747774566, | |
| "train_runtime": 7554.577, | |
| "train_samples_per_second": 508.301, | |
| "train_steps_per_second": 1.986 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 455, | |
| "total_flos": 1.7353653717551284e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |