| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.997356828193833, | |
| "global_step": 2830, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 9.307420494699646e-05, | |
| "loss": 0.5254, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.8732394576072693, | |
| "eval_loss": 0.30324122309684753, | |
| "eval_runtime": 39.6845, | |
| "eval_samples_per_second": 14.313, | |
| "eval_steps_per_second": 7.156, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 8.604240282685513e-05, | |
| "loss": 0.5153, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_accuracy": 0.9049295783042908, | |
| "eval_loss": 0.28999415040016174, | |
| "eval_runtime": 40.176, | |
| "eval_samples_per_second": 14.138, | |
| "eval_steps_per_second": 7.069, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 7.897526501766784e-05, | |
| "loss": 0.2832, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_accuracy": 0.9419013857841492, | |
| "eval_loss": 0.24485430121421814, | |
| "eval_runtime": 43.1522, | |
| "eval_samples_per_second": 13.163, | |
| "eval_steps_per_second": 6.581, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 7.190812720848057e-05, | |
| "loss": 0.2435, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_accuracy": 0.9436619877815247, | |
| "eval_loss": 0.18418200314044952, | |
| "eval_runtime": 40.2247, | |
| "eval_samples_per_second": 14.121, | |
| "eval_steps_per_second": 7.06, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 6.48409893992933e-05, | |
| "loss": 0.2056, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "eval_accuracy": 0.952464759349823, | |
| "eval_loss": 0.17370758950710297, | |
| "eval_runtime": 43.6775, | |
| "eval_samples_per_second": 13.004, | |
| "eval_steps_per_second": 6.502, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 5.7773851590106006e-05, | |
| "loss": 0.1485, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "eval_accuracy": 0.966549277305603, | |
| "eval_loss": 0.13336466252803802, | |
| "eval_runtime": 38.8727, | |
| "eval_samples_per_second": 14.612, | |
| "eval_steps_per_second": 7.306, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 5.0706713780918735e-05, | |
| "loss": 0.1717, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_accuracy": 0.9700704216957092, | |
| "eval_loss": 0.12312730401754379, | |
| "eval_runtime": 43.1241, | |
| "eval_samples_per_second": 13.171, | |
| "eval_steps_per_second": 6.586, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 4.367491166077738e-05, | |
| "loss": 0.1274, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "eval_accuracy": 0.9700704216957092, | |
| "eval_loss": 0.12308631092309952, | |
| "eval_runtime": 39.3033, | |
| "eval_samples_per_second": 14.452, | |
| "eval_steps_per_second": 7.226, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 3.660777385159011e-05, | |
| "loss": 0.1268, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "eval_accuracy": 0.9718309640884399, | |
| "eval_loss": 0.114546038210392, | |
| "eval_runtime": 38.8418, | |
| "eval_samples_per_second": 14.623, | |
| "eval_steps_per_second": 7.312, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 2.954063604240283e-05, | |
| "loss": 0.1395, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "eval_accuracy": 0.9771126508712769, | |
| "eval_loss": 0.09564723074436188, | |
| "eval_runtime": 41.3738, | |
| "eval_samples_per_second": 13.728, | |
| "eval_steps_per_second": 6.864, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "learning_rate": 2.247349823321555e-05, | |
| "loss": 0.0993, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "eval_accuracy": 0.98591548204422, | |
| "eval_loss": 0.05968547984957695, | |
| "eval_runtime": 41.7972, | |
| "eval_samples_per_second": 13.589, | |
| "eval_steps_per_second": 6.795, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 1.5406360424028267e-05, | |
| "loss": 0.0785, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "eval_accuracy": 0.9788732528686523, | |
| "eval_loss": 0.10253579914569855, | |
| "eval_runtime": 40.093, | |
| "eval_samples_per_second": 14.167, | |
| "eval_steps_per_second": 7.084, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 8.33922261484099e-06, | |
| "loss": 0.089, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "eval_accuracy": 0.9788732528686523, | |
| "eval_loss": 0.10597831010818481, | |
| "eval_runtime": 37.5593, | |
| "eval_samples_per_second": 15.123, | |
| "eval_steps_per_second": 7.561, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 1.2720848056537103e-06, | |
| "loss": 0.0796, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "eval_accuracy": 0.9806337952613831, | |
| "eval_loss": 0.09218709170818329, | |
| "eval_runtime": 40.4161, | |
| "eval_samples_per_second": 14.054, | |
| "eval_steps_per_second": 7.027, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2830, | |
| "total_flos": 1.2045174996823788e+18, | |
| "train_loss": 0.20077389284915723, | |
| "train_runtime": 3673.7626, | |
| "train_samples_per_second": 6.179, | |
| "train_steps_per_second": 0.77 | |
| } | |
| ], | |
| "max_steps": 2830, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.2045174996823788e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |