| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.9984, | |
| "global_step": 6240, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8989999890327454, | |
| "eval_loss": 0.2822728157043457, | |
| "eval_runtime": 4.6919, | |
| "eval_samples_per_second": 213.134, | |
| "eval_steps_per_second": 26.642, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.107842707402845e-05, | |
| "loss": 0.3203, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9089999794960022, | |
| "eval_loss": 0.3380243480205536, | |
| "eval_runtime": 4.8701, | |
| "eval_samples_per_second": 205.334, | |
| "eval_steps_per_second": 25.667, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9240000247955322, | |
| "eval_loss": 0.3920462727546692, | |
| "eval_runtime": 4.7219, | |
| "eval_samples_per_second": 211.777, | |
| "eval_steps_per_second": 26.472, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 2.8371247015315168e-05, | |
| "loss": 0.1019, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9200000166893005, | |
| "eval_loss": 0.47628748416900635, | |
| "eval_runtime": 4.8792, | |
| "eval_samples_per_second": 204.953, | |
| "eval_steps_per_second": 25.619, | |
| "step": 1248 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 2.5664066956601888e-05, | |
| "loss": 0.033, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9309999942779541, | |
| "eval_loss": 0.4591754674911499, | |
| "eval_runtime": 4.8857, | |
| "eval_samples_per_second": 204.677, | |
| "eval_steps_per_second": 25.585, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9160000085830688, | |
| "eval_loss": 0.555105447769165, | |
| "eval_runtime": 4.7722, | |
| "eval_samples_per_second": 209.547, | |
| "eval_steps_per_second": 26.193, | |
| "step": 1872 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 2.295688689788861e-05, | |
| "loss": 0.0122, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9200000166893005, | |
| "eval_loss": 0.6067560315132141, | |
| "eval_runtime": 4.772, | |
| "eval_samples_per_second": 209.555, | |
| "eval_steps_per_second": 26.194, | |
| "step": 2184 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9120000004768372, | |
| "eval_loss": 0.6876145601272583, | |
| "eval_runtime": 4.7132, | |
| "eval_samples_per_second": 212.168, | |
| "eval_steps_per_second": 26.521, | |
| "step": 2496 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 2.024970683917533e-05, | |
| "loss": 0.0158, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9129999876022339, | |
| "eval_loss": 0.7302923798561096, | |
| "eval_runtime": 4.7376, | |
| "eval_samples_per_second": 211.078, | |
| "eval_steps_per_second": 26.385, | |
| "step": 2808 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 1.754252678046205e-05, | |
| "loss": 0.0063, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9259999990463257, | |
| "eval_loss": 0.6355671286582947, | |
| "eval_runtime": 4.6719, | |
| "eval_samples_per_second": 214.047, | |
| "eval_steps_per_second": 26.756, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9210000038146973, | |
| "eval_loss": 0.6988717913627625, | |
| "eval_runtime": 4.8916, | |
| "eval_samples_per_second": 204.431, | |
| "eval_steps_per_second": 25.554, | |
| "step": 3432 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 1.483534672174877e-05, | |
| "loss": 0.0043, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9190000295639038, | |
| "eval_loss": 0.7033405900001526, | |
| "eval_runtime": 4.8604, | |
| "eval_samples_per_second": 205.746, | |
| "eval_steps_per_second": 25.718, | |
| "step": 3744 | |
| }, | |
| { | |
| "epoch": 12.82, | |
| "learning_rate": 1.2128166663035492e-05, | |
| "loss": 0.0047, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9070000052452087, | |
| "eval_loss": 0.840012788772583, | |
| "eval_runtime": 4.9023, | |
| "eval_samples_per_second": 203.986, | |
| "eval_steps_per_second": 25.498, | |
| "step": 4056 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9190000295639038, | |
| "eval_loss": 0.7390091419219971, | |
| "eval_runtime": 4.8503, | |
| "eval_samples_per_second": 206.175, | |
| "eval_steps_per_second": 25.772, | |
| "step": 4368 | |
| }, | |
| { | |
| "epoch": 14.42, | |
| "learning_rate": 9.420986604322212e-06, | |
| "loss": 0.0035, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9169999957084656, | |
| "eval_loss": 0.7360817790031433, | |
| "eval_runtime": 4.876, | |
| "eval_samples_per_second": 205.084, | |
| "eval_steps_per_second": 25.636, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9229999780654907, | |
| "eval_loss": 0.7316659688949585, | |
| "eval_runtime": 4.8948, | |
| "eval_samples_per_second": 204.299, | |
| "eval_steps_per_second": 25.537, | |
| "step": 4992 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "learning_rate": 6.713806545608932e-06, | |
| "loss": 0.002, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9139999747276306, | |
| "eval_loss": 0.8043374419212341, | |
| "eval_runtime": 4.6508, | |
| "eval_samples_per_second": 215.017, | |
| "eval_steps_per_second": 26.877, | |
| "step": 5304 | |
| }, | |
| { | |
| "epoch": 17.63, | |
| "learning_rate": 4.006626486895653e-06, | |
| "loss": 0.0, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9129999876022339, | |
| "eval_loss": 0.8095300197601318, | |
| "eval_runtime": 4.7921, | |
| "eval_samples_per_second": 208.677, | |
| "eval_steps_per_second": 26.085, | |
| "step": 5616 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9120000004768372, | |
| "eval_loss": 0.7782994508743286, | |
| "eval_runtime": 4.7356, | |
| "eval_samples_per_second": 211.168, | |
| "eval_steps_per_second": 26.396, | |
| "step": 5928 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "learning_rate": 1.2994464281823743e-06, | |
| "loss": 0.0007, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9139999747276306, | |
| "eval_loss": 0.7657350897789001, | |
| "eval_runtime": 4.8181, | |
| "eval_samples_per_second": 207.551, | |
| "eval_steps_per_second": 25.944, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 6240, | |
| "total_flos": 6577309216364544.0, | |
| "train_loss": 0.04045970406598197, | |
| "train_runtime": 2296.6324, | |
| "train_samples_per_second": 43.542, | |
| "train_steps_per_second": 2.717 | |
| } | |
| ], | |
| "max_steps": 6240, | |
| "num_train_epochs": 20, | |
| "total_flos": 6577309216364544.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |