| { |
| "best_metric": 56.2712800220852, |
| "best_model_checkpoint": "outputs/bert-tiny/weak_tiny_poe/qqp_42/checkpoint-11371", |
| "epoch": 3.0, |
| "global_step": 34113, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.04, |
| "learning_rate": 1.9706856623574592e-05, |
| "loss": 0.5147, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.09, |
| "learning_rate": 1.9413713247149183e-05, |
| "loss": 0.4998, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.13, |
| "learning_rate": 1.9120569870723774e-05, |
| "loss": 0.5112, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.18, |
| "learning_rate": 1.882742649429836e-05, |
| "loss": 0.506, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.22, |
| "learning_rate": 1.8534283117872952e-05, |
| "loss": 0.5112, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.26, |
| "learning_rate": 1.8241139741447543e-05, |
| "loss": 0.5002, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 1.7947996365022134e-05, |
| "loss": 0.4996, |
| "step": 3500 |
| }, |
| { |
| "epoch": 0.35, |
| "learning_rate": 1.7654852988596725e-05, |
| "loss": 0.5049, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.4, |
| "learning_rate": 1.7361709612171312e-05, |
| "loss": 0.5157, |
| "step": 4500 |
| }, |
| { |
| "epoch": 0.44, |
| "learning_rate": 1.7068566235745903e-05, |
| "loss": 0.5025, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.48, |
| "learning_rate": 1.6775422859320497e-05, |
| "loss": 0.4944, |
| "step": 5500 |
| }, |
| { |
| "epoch": 0.53, |
| "learning_rate": 1.6482279482895085e-05, |
| "loss": 0.5045, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.57, |
| "learning_rate": 1.6189136106469676e-05, |
| "loss": 0.4946, |
| "step": 6500 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 1.5895992730044266e-05, |
| "loss": 0.4964, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.66, |
| "learning_rate": 1.5602849353618857e-05, |
| "loss": 0.4944, |
| "step": 7500 |
| }, |
| { |
| "epoch": 0.7, |
| "learning_rate": 1.5309705977193448e-05, |
| "loss": 0.5124, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.75, |
| "learning_rate": 1.5016562600768037e-05, |
| "loss": 0.5155, |
| "step": 8500 |
| }, |
| { |
| "epoch": 0.79, |
| "learning_rate": 1.4723419224342628e-05, |
| "loss": 0.501, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.84, |
| "learning_rate": 1.4430275847917217e-05, |
| "loss": 0.4986, |
| "step": 9500 |
| }, |
| { |
| "epoch": 0.88, |
| "learning_rate": 1.4137132471491808e-05, |
| "loss": 0.4954, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.92, |
| "learning_rate": 1.3843989095066399e-05, |
| "loss": 0.4932, |
| "step": 10500 |
| }, |
| { |
| "epoch": 0.97, |
| "learning_rate": 1.3550845718640988e-05, |
| "loss": 0.4994, |
| "step": 11000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 52.985406876082116, |
| "eval_combined_score": 54.62834344908366, |
| "eval_f1": 56.2712800220852, |
| "eval_loss": 0.6511399745941162, |
| "eval_runtime": 91.3653, |
| "eval_samples_per_second": 442.509, |
| "eval_steps_per_second": 1.729, |
| "step": 11371 |
| }, |
| { |
| "epoch": 1.01, |
| "learning_rate": 1.3257702342215579e-05, |
| "loss": 0.5074, |
| "step": 11500 |
| }, |
| { |
| "epoch": 1.06, |
| "learning_rate": 1.2964558965790168e-05, |
| "loss": 0.4877, |
| "step": 12000 |
| }, |
| { |
| "epoch": 1.1, |
| "learning_rate": 1.2671415589364759e-05, |
| "loss": 0.5042, |
| "step": 12500 |
| }, |
| { |
| "epoch": 1.14, |
| "learning_rate": 1.2378272212939351e-05, |
| "loss": 0.4942, |
| "step": 13000 |
| }, |
| { |
| "epoch": 1.19, |
| "learning_rate": 1.208512883651394e-05, |
| "loss": 0.4809, |
| "step": 13500 |
| }, |
| { |
| "epoch": 1.23, |
| "learning_rate": 1.1791985460088531e-05, |
| "loss": 0.4964, |
| "step": 14000 |
| }, |
| { |
| "epoch": 1.28, |
| "learning_rate": 1.149884208366312e-05, |
| "loss": 0.5047, |
| "step": 14500 |
| }, |
| { |
| "epoch": 1.32, |
| "learning_rate": 1.1205698707237711e-05, |
| "loss": 0.503, |
| "step": 15000 |
| }, |
| { |
| "epoch": 1.36, |
| "learning_rate": 1.0912555330812302e-05, |
| "loss": 0.5007, |
| "step": 15500 |
| }, |
| { |
| "epoch": 1.41, |
| "learning_rate": 1.0619411954386891e-05, |
| "loss": 0.4955, |
| "step": 16000 |
| }, |
| { |
| "epoch": 1.45, |
| "learning_rate": 1.0326268577961482e-05, |
| "loss": 0.4932, |
| "step": 16500 |
| }, |
| { |
| "epoch": 1.5, |
| "learning_rate": 1.0033125201536071e-05, |
| "loss": 0.4962, |
| "step": 17000 |
| }, |
| { |
| "epoch": 1.54, |
| "learning_rate": 9.739981825110662e-06, |
| "loss": 0.4977, |
| "step": 17500 |
| }, |
| { |
| "epoch": 1.58, |
| "learning_rate": 9.446838448685253e-06, |
| "loss": 0.5096, |
| "step": 18000 |
| }, |
| { |
| "epoch": 1.63, |
| "learning_rate": 9.153695072259843e-06, |
| "loss": 0.498, |
| "step": 18500 |
| }, |
| { |
| "epoch": 1.67, |
| "learning_rate": 8.860551695834433e-06, |
| "loss": 0.4914, |
| "step": 19000 |
| }, |
| { |
| "epoch": 1.71, |
| "learning_rate": 8.567408319409023e-06, |
| "loss": 0.486, |
| "step": 19500 |
| }, |
| { |
| "epoch": 1.76, |
| "learning_rate": 8.274264942983614e-06, |
| "loss": 0.5059, |
| "step": 20000 |
| }, |
| { |
| "epoch": 1.8, |
| "learning_rate": 7.981121566558205e-06, |
| "loss": 0.4995, |
| "step": 20500 |
| }, |
| { |
| "epoch": 1.85, |
| "learning_rate": 7.687978190132794e-06, |
| "loss": 0.5073, |
| "step": 21000 |
| }, |
| { |
| "epoch": 1.89, |
| "learning_rate": 7.394834813707384e-06, |
| "loss": 0.5046, |
| "step": 21500 |
| }, |
| { |
| "epoch": 1.93, |
| "learning_rate": 7.101691437281976e-06, |
| "loss": 0.4898, |
| "step": 22000 |
| }, |
| { |
| "epoch": 1.98, |
| "learning_rate": 6.808548060856566e-06, |
| "loss": 0.5001, |
| "step": 22500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 55.87187731882266, |
| "eval_combined_score": 54.503487058371114, |
| "eval_f1": 53.135096797919566, |
| "eval_loss": 0.6367405652999878, |
| "eval_runtime": 76.8822, |
| "eval_samples_per_second": 525.869, |
| "eval_steps_per_second": 2.055, |
| "step": 22742 |
| }, |
| { |
| "epoch": 2.02, |
| "learning_rate": 6.515404684431156e-06, |
| "loss": 0.5015, |
| "step": 23000 |
| }, |
| { |
| "epoch": 2.07, |
| "learning_rate": 6.222261308005746e-06, |
| "loss": 0.4969, |
| "step": 23500 |
| }, |
| { |
| "epoch": 2.11, |
| "learning_rate": 5.929117931580336e-06, |
| "loss": 0.4855, |
| "step": 24000 |
| }, |
| { |
| "epoch": 2.15, |
| "learning_rate": 5.6359745551549274e-06, |
| "loss": 0.495, |
| "step": 24500 |
| }, |
| { |
| "epoch": 2.2, |
| "learning_rate": 5.3428311787295174e-06, |
| "loss": 0.4924, |
| "step": 25000 |
| }, |
| { |
| "epoch": 2.24, |
| "learning_rate": 5.049687802304107e-06, |
| "loss": 0.4938, |
| "step": 25500 |
| }, |
| { |
| "epoch": 2.29, |
| "learning_rate": 4.756544425878697e-06, |
| "loss": 0.4977, |
| "step": 26000 |
| }, |
| { |
| "epoch": 2.33, |
| "learning_rate": 4.463401049453288e-06, |
| "loss": 0.4904, |
| "step": 26500 |
| }, |
| { |
| "epoch": 2.37, |
| "learning_rate": 4.170257673027878e-06, |
| "loss": 0.4904, |
| "step": 27000 |
| }, |
| { |
| "epoch": 2.42, |
| "learning_rate": 3.877114296602469e-06, |
| "loss": 0.4802, |
| "step": 27500 |
| }, |
| { |
| "epoch": 2.46, |
| "learning_rate": 3.583970920177059e-06, |
| "loss": 0.4908, |
| "step": 28000 |
| }, |
| { |
| "epoch": 2.51, |
| "learning_rate": 3.290827543751649e-06, |
| "loss": 0.4951, |
| "step": 28500 |
| }, |
| { |
| "epoch": 2.55, |
| "learning_rate": 2.9976841673262398e-06, |
| "loss": 0.4942, |
| "step": 29000 |
| }, |
| { |
| "epoch": 2.59, |
| "learning_rate": 2.7045407909008298e-06, |
| "loss": 0.5074, |
| "step": 29500 |
| }, |
| { |
| "epoch": 2.64, |
| "learning_rate": 2.41139741447542e-06, |
| "loss": 0.4894, |
| "step": 30000 |
| }, |
| { |
| "epoch": 2.68, |
| "learning_rate": 2.1182540380500106e-06, |
| "loss": 0.486, |
| "step": 30500 |
| }, |
| { |
| "epoch": 2.73, |
| "learning_rate": 1.825110661624601e-06, |
| "loss": 0.4878, |
| "step": 31000 |
| }, |
| { |
| "epoch": 2.77, |
| "learning_rate": 1.531967285199191e-06, |
| "loss": 0.4943, |
| "step": 31500 |
| }, |
| { |
| "epoch": 2.81, |
| "learning_rate": 1.2388239087737813e-06, |
| "loss": 0.499, |
| "step": 32000 |
| }, |
| { |
| "epoch": 2.86, |
| "learning_rate": 9.456805323483717e-07, |
| "loss": 0.4934, |
| "step": 32500 |
| }, |
| { |
| "epoch": 2.9, |
| "learning_rate": 6.525371559229619e-07, |
| "loss": 0.4995, |
| "step": 33000 |
| }, |
| { |
| "epoch": 2.95, |
| "learning_rate": 3.5939377949755233e-07, |
| "loss": 0.4977, |
| "step": 33500 |
| }, |
| { |
| "epoch": 2.99, |
| "learning_rate": 6.62504030721426e-08, |
| "loss": 0.5048, |
| "step": 34000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 57.35097699727925, |
| "eval_combined_score": 54.23605203862318, |
| "eval_f1": 51.12112707996711, |
| "eval_loss": 0.6283207535743713, |
| "eval_runtime": 75.3789, |
| "eval_samples_per_second": 536.357, |
| "eval_steps_per_second": 2.096, |
| "step": 34113 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 34113, |
| "total_flos": 0.0, |
| "train_loss": 0.4982893261680103, |
| "train_runtime": 2110.3134, |
| "train_samples_per_second": 517.24, |
| "train_steps_per_second": 16.165 |
| } |
| ], |
| "max_steps": 34113, |
| "num_train_epochs": 3, |
| "total_flos": 0.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
