| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 20276, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.8767015190372854e-05, | |
| "loss": 3.6543, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.7534030380745706e-05, | |
| "loss": 2.9884, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.630104557111857e-05, | |
| "loss": 2.6586, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.506806076149142e-05, | |
| "loss": 2.6187, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.3835075951864274e-05, | |
| "loss": 2.5105, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.2602091142237125e-05, | |
| "loss": 2.235, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.1369106332609984e-05, | |
| "loss": 2.2836, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.013612152298284e-05, | |
| "loss": 2.1538, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.8903136713355694e-05, | |
| "loss": 2.2132, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.7670151903728545e-05, | |
| "loss": 2.1561, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_e": 0.4684004474272931, | |
| "eval_f1": 0.4364628001202054, | |
| "eval_loss": 3.77375864982605, | |
| "eval_runtime": 80.8413, | |
| "eval_samples_per_second": 44.235, | |
| "eval_steps_per_second": 44.235, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.6437167094101404e-05, | |
| "loss": 2.0688, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 3.5204182284474255e-05, | |
| "loss": 2.0712, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.3971197474847113e-05, | |
| "loss": 1.7271, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 3.2738212665219965e-05, | |
| "loss": 1.7146, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 3.150522785559282e-05, | |
| "loss": 1.8106, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.0272243045965675e-05, | |
| "loss": 1.7478, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 2.903925823633853e-05, | |
| "loss": 1.8638, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 2.780627342671138e-05, | |
| "loss": 1.41, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.657328861708424e-05, | |
| "loss": 1.7941, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 2.5340303807457095e-05, | |
| "loss": 1.5219, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_e": 0.5542505592841164, | |
| "eval_f1": 0.5065431932934646, | |
| "eval_loss": 2.430802822113037, | |
| "eval_runtime": 80.8808, | |
| "eval_samples_per_second": 44.213, | |
| "eval_steps_per_second": 44.213, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 2.4107318997829946e-05, | |
| "loss": 1.2346, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 2.2874334188202805e-05, | |
| "loss": 1.2315, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 2.1641349378575656e-05, | |
| "loss": 1.2655, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 2.040836456894851e-05, | |
| "loss": 1.2839, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.9175379759321366e-05, | |
| "loss": 1.0951, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.794239494969422e-05, | |
| "loss": 1.2722, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.6709410140067076e-05, | |
| "loss": 1.0126, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.547642533043993e-05, | |
| "loss": 1.1157, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.4243440520812784e-05, | |
| "loss": 1.1724, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.301045571118564e-05, | |
| "loss": 0.9838, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_e": 0.5724272930648769, | |
| "eval_f1": 0.5272328806773431, | |
| "eval_loss": 3.077216148376465, | |
| "eval_runtime": 81.057, | |
| "eval_samples_per_second": 44.117, | |
| "eval_steps_per_second": 44.117, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 1.1777470901558493e-05, | |
| "loss": 0.9351, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 1.0544486091931348e-05, | |
| "loss": 0.997, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.311501282304202e-06, | |
| "loss": 0.9782, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.078516472677057e-06, | |
| "loss": 0.9934, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 6.8455316630499115e-06, | |
| "loss": 0.9395, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 5.6125468534227665e-06, | |
| "loss": 0.8001, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.379562043795621e-06, | |
| "loss": 0.9129, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.1465772341684756e-06, | |
| "loss": 1.0251, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.9135924245413297e-06, | |
| "loss": 1.0114, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 6.806076149141843e-07, | |
| "loss": 0.8334, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_e": 0.6009507829977628, | |
| "eval_f1": 0.5501093051292324, | |
| "eval_loss": 2.754518747329712, | |
| "eval_runtime": 81.0315, | |
| "eval_samples_per_second": 44.131, | |
| "eval_steps_per_second": 44.131, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 20276, | |
| "total_flos": 1.023741888632904e+16, | |
| "train_loss": 1.5977188166040575, | |
| "train_runtime": 2455.1123, | |
| "train_samples_per_second": 8.259, | |
| "train_steps_per_second": 8.259 | |
| } | |
| ], | |
| "max_steps": 20276, | |
| "num_train_epochs": 2, | |
| "total_flos": 1.023741888632904e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |