| { | |
| "best_metric": 0.0029988216701895, | |
| "best_model_checkpoint": "./models/day-night/checkpoint-800", | |
| "epoch": 5.0, | |
| "global_step": 1680, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 9.410714285714286e-05, | |
| "loss": 0.2756, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.815476190476191e-05, | |
| "loss": 0.1021, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.9734411085450346, | |
| "eval_loss": 0.06794659793376923, | |
| "eval_runtime": 7.5982, | |
| "eval_samples_per_second": 113.974, | |
| "eval_steps_per_second": 14.345, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 8.226190476190476e-05, | |
| "loss": 0.1375, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 7.630952380952381e-05, | |
| "loss": 0.0199, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_accuracy": 0.9919168591224018, | |
| "eval_loss": 0.018402865156531334, | |
| "eval_runtime": 7.4284, | |
| "eval_samples_per_second": 116.58, | |
| "eval_steps_per_second": 14.673, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 7.035714285714287e-05, | |
| "loss": 0.0518, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 6.446428571428572e-05, | |
| "loss": 0.0723, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_accuracy": 0.7852193995381063, | |
| "eval_loss": 0.6625365018844604, | |
| "eval_runtime": 7.266, | |
| "eval_samples_per_second": 119.184, | |
| "eval_steps_per_second": 15.001, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 5.851190476190477e-05, | |
| "loss": 0.0278, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 5.2559523809523814e-05, | |
| "loss": 0.0247, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "eval_accuracy": 0.9988452655889145, | |
| "eval_loss": 0.0029988216701895, | |
| "eval_runtime": 7.7626, | |
| "eval_samples_per_second": 111.56, | |
| "eval_steps_per_second": 14.042, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 4.660714285714286e-05, | |
| "loss": 0.0576, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 4.0654761904761904e-05, | |
| "loss": 0.0273, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.9884526558891455, | |
| "eval_loss": 0.025376001372933388, | |
| "eval_runtime": 7.4049, | |
| "eval_samples_per_second": 116.95, | |
| "eval_steps_per_second": 14.72, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 3.4702380952380956e-05, | |
| "loss": 0.0116, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 2.8749999999999997e-05, | |
| "loss": 0.012, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "eval_accuracy": 0.9965357967667436, | |
| "eval_loss": 0.017664099112153053, | |
| "eval_runtime": 7.9179, | |
| "eval_samples_per_second": 109.373, | |
| "eval_steps_per_second": 13.766, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 2.279761904761905e-05, | |
| "loss": 0.0266, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 1.6845238095238097e-05, | |
| "loss": 0.0142, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "eval_accuracy": 0.9907621247113164, | |
| "eval_loss": 0.028172077611088753, | |
| "eval_runtime": 7.4701, | |
| "eval_samples_per_second": 115.929, | |
| "eval_steps_per_second": 14.591, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 1.0892857142857144e-05, | |
| "loss": 0.0157, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 4.940476190476191e-06, | |
| "loss": 0.009, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "eval_accuracy": 0.9976905311778291, | |
| "eval_loss": 0.01886291243135929, | |
| "eval_runtime": 7.2095, | |
| "eval_samples_per_second": 120.119, | |
| "eval_steps_per_second": 15.119, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 1680, | |
| "total_flos": 1.0500251940864e+16, | |
| "train_loss": 0.05322713415537562, | |
| "train_runtime": 260.1266, | |
| "train_samples_per_second": 64.507, | |
| "train_steps_per_second": 6.458 | |
| } | |
| ], | |
| "max_steps": 1680, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.0500251940864e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |