| { | |
| "best_metric": 0.8910891089108911, | |
| "best_model_checkpoint": "Electrcical-IMAGE-finetuned-eurosat/checkpoint-199", | |
| "epoch": 9.824561403508772, | |
| "eval_steps": 500, | |
| "global_step": 280, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3508771929824561, | |
| "grad_norm": 10.464795112609863, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.7823, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "grad_norm": 9.879911422729492, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.4879, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9824561403508771, | |
| "eval_accuracy": 0.7326732673267327, | |
| "eval_loss": 0.9158226847648621, | |
| "eval_runtime": 116.3827, | |
| "eval_samples_per_second": 3.471, | |
| "eval_steps_per_second": 0.112, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 7.517858028411865, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 1.1987, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.4035087719298245, | |
| "grad_norm": 7.376903533935547, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.8925, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.7543859649122808, | |
| "grad_norm": 8.916149139404297, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 0.7072, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8366336633663366, | |
| "eval_loss": 0.4648021161556244, | |
| "eval_runtime": 4.3367, | |
| "eval_samples_per_second": 93.158, | |
| "eval_steps_per_second": 2.998, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 8.09045124053955, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.665, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.456140350877193, | |
| "grad_norm": 9.807299613952637, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.5713, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.807017543859649, | |
| "grad_norm": 10.828336715698242, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.521, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.982456140350877, | |
| "eval_accuracy": 0.8712871287128713, | |
| "eval_loss": 0.3816491961479187, | |
| "eval_runtime": 4.1894, | |
| "eval_samples_per_second": 96.434, | |
| "eval_steps_per_second": 3.103, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.1578947368421053, | |
| "grad_norm": 6.668788433074951, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 0.5152, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5087719298245617, | |
| "grad_norm": 5.836483478546143, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.473, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.8596491228070176, | |
| "grad_norm": 5.453497409820557, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.4664, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8564356435643564, | |
| "eval_loss": 0.40334010124206543, | |
| "eval_runtime": 4.398, | |
| "eval_samples_per_second": 91.86, | |
| "eval_steps_per_second": 2.956, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.2105263157894735, | |
| "grad_norm": 6.095137596130371, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.3831, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.56140350877193, | |
| "grad_norm": 6.058220386505127, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.4376, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.912280701754386, | |
| "grad_norm": 6.937771797180176, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.3944, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.982456140350877, | |
| "eval_accuracy": 0.8737623762376238, | |
| "eval_loss": 0.3690718114376068, | |
| "eval_runtime": 4.2858, | |
| "eval_samples_per_second": 94.264, | |
| "eval_steps_per_second": 3.033, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 5.2631578947368425, | |
| "grad_norm": 6.101373672485352, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.3582, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.614035087719298, | |
| "grad_norm": 8.85653305053711, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.3584, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.964912280701754, | |
| "grad_norm": 7.283915996551514, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.3627, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8886138613861386, | |
| "eval_loss": 0.3214375674724579, | |
| "eval_runtime": 4.5245, | |
| "eval_samples_per_second": 89.292, | |
| "eval_steps_per_second": 2.873, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 6.315789473684211, | |
| "grad_norm": 5.079178810119629, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.3304, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 5.342247486114502, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.3298, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.982456140350877, | |
| "eval_accuracy": 0.8910891089108911, | |
| "eval_loss": 0.3172283470630646, | |
| "eval_runtime": 4.1397, | |
| "eval_samples_per_second": 97.593, | |
| "eval_steps_per_second": 3.14, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 7.017543859649122, | |
| "grad_norm": 6.236889362335205, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.342, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.368421052631579, | |
| "grad_norm": 9.212471008300781, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.3288, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 7.719298245614035, | |
| "grad_norm": 5.810153484344482, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.3203, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8910891089108911, | |
| "eval_loss": 0.3060537278652191, | |
| "eval_runtime": 4.5988, | |
| "eval_samples_per_second": 87.848, | |
| "eval_steps_per_second": 2.827, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 8.070175438596491, | |
| "grad_norm": 5.650562763214111, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.3379, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 8.421052631578947, | |
| "grad_norm": 3.9896435737609863, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.2913, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 8.771929824561404, | |
| "grad_norm": 7.3288397789001465, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.2737, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.982456140350877, | |
| "eval_accuracy": 0.8861386138613861, | |
| "eval_loss": 0.3128886818885803, | |
| "eval_runtime": 4.2212, | |
| "eval_samples_per_second": 95.706, | |
| "eval_steps_per_second": 3.08, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 9.12280701754386, | |
| "grad_norm": 5.5674285888671875, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.2991, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 9.473684210526315, | |
| "grad_norm": 6.089910984039307, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.2943, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 9.824561403508772, | |
| "grad_norm": 5.232763290405273, | |
| "learning_rate": 0.0, | |
| "loss": 0.2728, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 9.824561403508772, | |
| "eval_accuracy": 0.8861386138613861, | |
| "eval_loss": 0.308758020401001, | |
| "eval_runtime": 4.3884, | |
| "eval_samples_per_second": 92.061, | |
| "eval_steps_per_second": 2.962, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 9.824561403508772, | |
| "step": 280, | |
| "total_flos": 8.878060093031055e+17, | |
| "train_loss": 0.5355421313217708, | |
| "train_runtime": 1702.5623, | |
| "train_samples_per_second": 21.344, | |
| "train_steps_per_second": 0.164 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 280, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 8.878060093031055e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |