| { |
| "best_metric": 0.7684563758389261, |
| "best_model_checkpoint": "./drive/Shareddrives/CS198-Drones/[v5] Training Output/vit-base-patch16-224_rice-leaf-disease-augmented-v4_v5_pft/checkpoint-3392", |
| "epoch": 30.0, |
| "eval_steps": 64, |
| "global_step": 3840, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.5, |
| "grad_norm": 3.3553061485290527, |
| "learning_rate": 1.25e-05, |
| "loss": 2.1933, |
| "step": 64 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_accuracy": 0.09060402684563758, |
| "eval_loss": 2.1968395709991455, |
| "eval_runtime": 9.159, |
| "eval_samples_per_second": 32.536, |
| "eval_steps_per_second": 0.546, |
| "step": 64 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 3.834850311279297, |
| "learning_rate": 2.5e-05, |
| "loss": 2.1246, |
| "step": 128 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.12416107382550336, |
| "eval_loss": 2.108480453491211, |
| "eval_runtime": 8.2948, |
| "eval_samples_per_second": 35.926, |
| "eval_steps_per_second": 0.603, |
| "step": 128 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 3.4604074954986572, |
| "learning_rate": 3.7500000000000003e-05, |
| "loss": 2.021, |
| "step": 192 |
| }, |
| { |
| "epoch": 1.5, |
| "eval_accuracy": 0.2181208053691275, |
| "eval_loss": 1.9747371673583984, |
| "eval_runtime": 8.8968, |
| "eval_samples_per_second": 33.495, |
| "eval_steps_per_second": 0.562, |
| "step": 192 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 3.332814931869507, |
| "learning_rate": 5e-05, |
| "loss": 1.8907, |
| "step": 256 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.3288590604026846, |
| "eval_loss": 1.8124574422836304, |
| "eval_runtime": 7.9216, |
| "eval_samples_per_second": 37.619, |
| "eval_steps_per_second": 0.631, |
| "step": 256 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 2.982445478439331, |
| "learning_rate": 4.8597083257709194e-05, |
| "loss": 1.7317, |
| "step": 320 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_accuracy": 0.39932885906040266, |
| "eval_loss": 1.6570496559143066, |
| "eval_runtime": 7.7262, |
| "eval_samples_per_second": 38.57, |
| "eval_steps_per_second": 0.647, |
| "step": 320 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 2.638735771179199, |
| "learning_rate": 4.454578706170075e-05, |
| "loss": 1.614, |
| "step": 384 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.4798657718120805, |
| "eval_loss": 1.5440526008605957, |
| "eval_runtime": 7.8344, |
| "eval_samples_per_second": 38.038, |
| "eval_steps_per_second": 0.638, |
| "step": 384 |
| }, |
| { |
| "epoch": 3.5, |
| "grad_norm": 2.8015382289886475, |
| "learning_rate": 3.830080191288342e-05, |
| "loss": 1.518, |
| "step": 448 |
| }, |
| { |
| "epoch": 3.5, |
| "eval_accuracy": 0.5100671140939598, |
| "eval_loss": 1.4556909799575806, |
| "eval_runtime": 7.7497, |
| "eval_samples_per_second": 38.453, |
| "eval_steps_per_second": 0.645, |
| "step": 448 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 2.9463720321655273, |
| "learning_rate": 3.056302334890786e-05, |
| "loss": 1.4343, |
| "step": 512 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.5369127516778524, |
| "eval_loss": 1.3944100141525269, |
| "eval_runtime": 8.033, |
| "eval_samples_per_second": 37.097, |
| "eval_steps_per_second": 0.622, |
| "step": 512 |
| }, |
| { |
| "epoch": 4.5, |
| "grad_norm": 3.31219744682312, |
| "learning_rate": 2.2200888097417307e-05, |
| "loss": 1.4007, |
| "step": 576 |
| }, |
| { |
| "epoch": 4.5, |
| "eval_accuracy": 0.5469798657718121, |
| "eval_loss": 1.3522661924362183, |
| "eval_runtime": 7.6521, |
| "eval_samples_per_second": 38.943, |
| "eval_steps_per_second": 0.653, |
| "step": 576 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 2.2543766498565674, |
| "learning_rate": 1.4152906522061048e-05, |
| "loss": 1.345, |
| "step": 640 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.5536912751677853, |
| "eval_loss": 1.3297712802886963, |
| "eval_runtime": 7.6521, |
| "eval_samples_per_second": 38.944, |
| "eval_steps_per_second": 0.653, |
| "step": 640 |
| }, |
| { |
| "epoch": 5.5, |
| "grad_norm": 2.3809094429016113, |
| "learning_rate": 7.3223304703363135e-06, |
| "loss": 1.3353, |
| "step": 704 |
| }, |
| { |
| "epoch": 5.5, |
| "eval_accuracy": 0.5604026845637584, |
| "eval_loss": 1.3163374662399292, |
| "eval_runtime": 7.8107, |
| "eval_samples_per_second": 38.153, |
| "eval_steps_per_second": 0.64, |
| "step": 704 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.5822391510009766, |
| "learning_rate": 2.475778302439524e-06, |
| "loss": 1.324, |
| "step": 768 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.5604026845637584, |
| "eval_loss": 1.3087968826293945, |
| "eval_runtime": 8.6399, |
| "eval_samples_per_second": 34.491, |
| "eval_steps_per_second": 0.579, |
| "step": 768 |
| }, |
| { |
| "epoch": 6.5, |
| "grad_norm": 2.5432515144348145, |
| "learning_rate": 1.571947526689349e-07, |
| "loss": 1.3386, |
| "step": 832 |
| }, |
| { |
| "epoch": 6.5, |
| "eval_accuracy": 0.5604026845637584, |
| "eval_loss": 1.3072452545166016, |
| "eval_runtime": 8.7139, |
| "eval_samples_per_second": 34.198, |
| "eval_steps_per_second": 0.574, |
| "step": 832 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 2.583087682723999, |
| "learning_rate": 4.937319780454559e-05, |
| "loss": 1.2882, |
| "step": 896 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.610738255033557, |
| "eval_loss": 1.2582257986068726, |
| "eval_runtime": 8.6431, |
| "eval_samples_per_second": 34.478, |
| "eval_steps_per_second": 0.578, |
| "step": 896 |
| }, |
| { |
| "epoch": 7.5, |
| "grad_norm": 2.8682312965393066, |
| "learning_rate": 4.6168104980707107e-05, |
| "loss": 1.2525, |
| "step": 960 |
| }, |
| { |
| "epoch": 7.5, |
| "eval_accuracy": 0.610738255033557, |
| "eval_loss": 1.2107785940170288, |
| "eval_runtime": 8.7628, |
| "eval_samples_per_second": 34.007, |
| "eval_steps_per_second": 0.571, |
| "step": 960 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 2.307509660720825, |
| "learning_rate": 4.058724504646835e-05, |
| "loss": 1.2039, |
| "step": 1024 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.6409395973154363, |
| "eval_loss": 1.1605823040008545, |
| "eval_runtime": 8.7427, |
| "eval_samples_per_second": 34.085, |
| "eval_steps_per_second": 0.572, |
| "step": 1024 |
| }, |
| { |
| "epoch": 8.5, |
| "grad_norm": 2.0223543643951416, |
| "learning_rate": 3.3256976548879184e-05, |
| "loss": 1.16, |
| "step": 1088 |
| }, |
| { |
| "epoch": 8.5, |
| "eval_accuracy": 0.6409395973154363, |
| "eval_loss": 1.126598834991455, |
| "eval_runtime": 8.7703, |
| "eval_samples_per_second": 33.979, |
| "eval_steps_per_second": 0.57, |
| "step": 1088 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 2.199993848800659, |
| "learning_rate": 2.5e-05, |
| "loss": 1.1401, |
| "step": 1152 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.6610738255033557, |
| "eval_loss": 1.1063629388809204, |
| "eval_runtime": 7.9893, |
| "eval_samples_per_second": 37.3, |
| "eval_steps_per_second": 0.626, |
| "step": 1152 |
| }, |
| { |
| "epoch": 9.5, |
| "grad_norm": 2.102375030517578, |
| "learning_rate": 1.6743023451120822e-05, |
| "loss": 1.1161, |
| "step": 1216 |
| }, |
| { |
| "epoch": 9.5, |
| "eval_accuracy": 0.6677852348993288, |
| "eval_loss": 1.0873396396636963, |
| "eval_runtime": 8.14, |
| "eval_samples_per_second": 36.609, |
| "eval_steps_per_second": 0.614, |
| "step": 1216 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 2.628359794616699, |
| "learning_rate": 9.412754953531672e-06, |
| "loss": 1.0979, |
| "step": 1280 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.6677852348993288, |
| "eval_loss": 1.078514575958252, |
| "eval_runtime": 8.8572, |
| "eval_samples_per_second": 33.645, |
| "eval_steps_per_second": 0.565, |
| "step": 1280 |
| }, |
| { |
| "epoch": 10.5, |
| "grad_norm": 2.5552382469177246, |
| "learning_rate": 3.831895019292909e-06, |
| "loss": 1.0961, |
| "step": 1344 |
| }, |
| { |
| "epoch": 10.5, |
| "eval_accuracy": 0.6644295302013423, |
| "eval_loss": 1.0734963417053223, |
| "eval_runtime": 8.8197, |
| "eval_samples_per_second": 33.788, |
| "eval_steps_per_second": 0.567, |
| "step": 1344 |
| }, |
| { |
| "epoch": 11.0, |
| "grad_norm": 2.206542491912842, |
| "learning_rate": 6.268021954544068e-07, |
| "loss": 1.0852, |
| "step": 1408 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_accuracy": 0.6677852348993288, |
| "eval_loss": 1.0718457698822021, |
| "eval_runtime": 9.1051, |
| "eval_samples_per_second": 32.729, |
| "eval_steps_per_second": 0.549, |
| "step": 1408 |
| }, |
| { |
| "epoch": 11.5, |
| "grad_norm": 3.107128381729126, |
| "learning_rate": 4.984280524733107e-05, |
| "loss": 1.0845, |
| "step": 1472 |
| }, |
| { |
| "epoch": 11.5, |
| "eval_accuracy": 0.6711409395973155, |
| "eval_loss": 1.0672944784164429, |
| "eval_runtime": 8.7733, |
| "eval_samples_per_second": 33.967, |
| "eval_steps_per_second": 0.57, |
| "step": 1472 |
| }, |
| { |
| "epoch": 12.0, |
| "grad_norm": 2.55963397026062, |
| "learning_rate": 4.7524221697560484e-05, |
| "loss": 1.0652, |
| "step": 1536 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_accuracy": 0.697986577181208, |
| "eval_loss": 1.019176959991455, |
| "eval_runtime": 7.8711, |
| "eval_samples_per_second": 37.86, |
| "eval_steps_per_second": 0.635, |
| "step": 1536 |
| }, |
| { |
| "epoch": 12.5, |
| "grad_norm": 2.3591487407684326, |
| "learning_rate": 4.267766952966369e-05, |
| "loss": 1.0336, |
| "step": 1600 |
| }, |
| { |
| "epoch": 12.5, |
| "eval_accuracy": 0.6946308724832215, |
| "eval_loss": 0.9978868961334229, |
| "eval_runtime": 8.7958, |
| "eval_samples_per_second": 33.88, |
| "eval_steps_per_second": 0.568, |
| "step": 1600 |
| }, |
| { |
| "epoch": 13.0, |
| "grad_norm": 2.140841484069824, |
| "learning_rate": 3.5847093477938956e-05, |
| "loss": 1.006, |
| "step": 1664 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_accuracy": 0.697986577181208, |
| "eval_loss": 0.9768925905227661, |
| "eval_runtime": 9.0047, |
| "eval_samples_per_second": 33.094, |
| "eval_steps_per_second": 0.555, |
| "step": 1664 |
| }, |
| { |
| "epoch": 13.5, |
| "grad_norm": 2.3379549980163574, |
| "learning_rate": 2.7799111902582686e-05, |
| "loss": 0.9842, |
| "step": 1728 |
| }, |
| { |
| "epoch": 13.5, |
| "eval_accuracy": 0.7013422818791947, |
| "eval_loss": 0.9594218730926514, |
| "eval_runtime": 8.8056, |
| "eval_samples_per_second": 33.842, |
| "eval_steps_per_second": 0.568, |
| "step": 1728 |
| }, |
| { |
| "epoch": 14.0, |
| "grad_norm": 2.195948362350464, |
| "learning_rate": 1.9436976651092164e-05, |
| "loss": 0.9735, |
| "step": 1792 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_accuracy": 0.7080536912751678, |
| "eval_loss": 0.9499163627624512, |
| "eval_runtime": 8.8229, |
| "eval_samples_per_second": 33.776, |
| "eval_steps_per_second": 0.567, |
| "step": 1792 |
| }, |
| { |
| "epoch": 14.5, |
| "grad_norm": 2.181391716003418, |
| "learning_rate": 1.169919808711657e-05, |
| "loss": 0.9662, |
| "step": 1856 |
| }, |
| { |
| "epoch": 14.5, |
| "eval_accuracy": 0.7114093959731543, |
| "eval_loss": 0.9434810280799866, |
| "eval_runtime": 8.7331, |
| "eval_samples_per_second": 34.123, |
| "eval_steps_per_second": 0.573, |
| "step": 1856 |
| }, |
| { |
| "epoch": 15.0, |
| "grad_norm": 1.8749948740005493, |
| "learning_rate": 5.454212938299264e-06, |
| "loss": 0.9491, |
| "step": 1920 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_accuracy": 0.714765100671141, |
| "eval_loss": 0.9374688863754272, |
| "eval_runtime": 8.9079, |
| "eval_samples_per_second": 33.453, |
| "eval_steps_per_second": 0.561, |
| "step": 1920 |
| }, |
| { |
| "epoch": 15.5, |
| "grad_norm": 2.289724111557007, |
| "learning_rate": 1.4029167422908107e-06, |
| "loss": 0.945, |
| "step": 1984 |
| }, |
| { |
| "epoch": 15.5, |
| "eval_accuracy": 0.7181208053691275, |
| "eval_loss": 0.9357881546020508, |
| "eval_runtime": 8.649, |
| "eval_samples_per_second": 34.455, |
| "eval_steps_per_second": 0.578, |
| "step": 1984 |
| }, |
| { |
| "epoch": 16.0, |
| "grad_norm": 1.9242281913757324, |
| "learning_rate": 5e-05, |
| "loss": 0.9564, |
| "step": 2048 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_accuracy": 0.7181208053691275, |
| "eval_loss": 0.9354940056800842, |
| "eval_runtime": 8.9028, |
| "eval_samples_per_second": 33.473, |
| "eval_steps_per_second": 0.562, |
| "step": 2048 |
| }, |
| { |
| "epoch": 16.5, |
| "grad_norm": 2.045414924621582, |
| "learning_rate": 4.8597083257709174e-05, |
| "loss": 0.9448, |
| "step": 2112 |
| }, |
| { |
| "epoch": 16.5, |
| "eval_accuracy": 0.7248322147651006, |
| "eval_loss": 0.9101002216339111, |
| "eval_runtime": 8.8831, |
| "eval_samples_per_second": 33.547, |
| "eval_steps_per_second": 0.563, |
| "step": 2112 |
| }, |
| { |
| "epoch": 17.0, |
| "grad_norm": 1.940545916557312, |
| "learning_rate": 4.454578706170074e-05, |
| "loss": 0.9156, |
| "step": 2176 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_accuracy": 0.7214765100671141, |
| "eval_loss": 0.8984754085540771, |
| "eval_runtime": 8.7245, |
| "eval_samples_per_second": 34.157, |
| "eval_steps_per_second": 0.573, |
| "step": 2176 |
| }, |
| { |
| "epoch": 17.5, |
| "grad_norm": 2.2112772464752197, |
| "learning_rate": 3.8300801912883396e-05, |
| "loss": 0.9164, |
| "step": 2240 |
| }, |
| { |
| "epoch": 17.5, |
| "eval_accuracy": 0.7348993288590604, |
| "eval_loss": 0.878176212310791, |
| "eval_runtime": 8.8341, |
| "eval_samples_per_second": 33.733, |
| "eval_steps_per_second": 0.566, |
| "step": 2240 |
| }, |
| { |
| "epoch": 18.0, |
| "grad_norm": 1.7843997478485107, |
| "learning_rate": 3.056302334890788e-05, |
| "loss": 0.8709, |
| "step": 2304 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_accuracy": 0.7348993288590604, |
| "eval_loss": 0.86415696144104, |
| "eval_runtime": 8.1554, |
| "eval_samples_per_second": 36.54, |
| "eval_steps_per_second": 0.613, |
| "step": 2304 |
| }, |
| { |
| "epoch": 18.5, |
| "grad_norm": 2.0073421001434326, |
| "learning_rate": 2.2200888097417317e-05, |
| "loss": 0.8732, |
| "step": 2368 |
| }, |
| { |
| "epoch": 18.5, |
| "eval_accuracy": 0.7348993288590604, |
| "eval_loss": 0.8535670042037964, |
| "eval_runtime": 7.8654, |
| "eval_samples_per_second": 37.888, |
| "eval_steps_per_second": 0.636, |
| "step": 2368 |
| }, |
| { |
| "epoch": 19.0, |
| "grad_norm": 2.097783327102661, |
| "learning_rate": 1.4152906522061087e-05, |
| "loss": 0.8662, |
| "step": 2432 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_accuracy": 0.7449664429530202, |
| "eval_loss": 0.8506665229797363, |
| "eval_runtime": 8.1216, |
| "eval_samples_per_second": 36.692, |
| "eval_steps_per_second": 0.616, |
| "step": 2432 |
| }, |
| { |
| "epoch": 19.5, |
| "grad_norm": 2.2401092052459717, |
| "learning_rate": 7.3223304703363135e-06, |
| "loss": 0.8561, |
| "step": 2496 |
| }, |
| { |
| "epoch": 19.5, |
| "eval_accuracy": 0.7416107382550335, |
| "eval_loss": 0.8498286008834839, |
| "eval_runtime": 8.845, |
| "eval_samples_per_second": 33.691, |
| "eval_steps_per_second": 0.565, |
| "step": 2496 |
| }, |
| { |
| "epoch": 20.0, |
| "grad_norm": 2.1345531940460205, |
| "learning_rate": 2.475778302439505e-06, |
| "loss": 0.8608, |
| "step": 2560 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_accuracy": 0.7449664429530202, |
| "eval_loss": 0.8466877341270447, |
| "eval_runtime": 8.7517, |
| "eval_samples_per_second": 34.051, |
| "eval_steps_per_second": 0.571, |
| "step": 2560 |
| }, |
| { |
| "epoch": 20.5, |
| "grad_norm": 1.8704792261123657, |
| "learning_rate": 1.571947526689349e-07, |
| "loss": 0.8638, |
| "step": 2624 |
| }, |
| { |
| "epoch": 20.5, |
| "eval_accuracy": 0.7449664429530202, |
| "eval_loss": 0.8461165428161621, |
| "eval_runtime": 8.6825, |
| "eval_samples_per_second": 34.322, |
| "eval_steps_per_second": 0.576, |
| "step": 2624 |
| }, |
| { |
| "epoch": 21.0, |
| "grad_norm": 2.0402615070343018, |
| "learning_rate": 4.93731978045456e-05, |
| "loss": 0.8449, |
| "step": 2688 |
| }, |
| { |
| "epoch": 21.0, |
| "eval_accuracy": 0.7483221476510067, |
| "eval_loss": 0.8376011252403259, |
| "eval_runtime": 8.1707, |
| "eval_samples_per_second": 36.472, |
| "eval_steps_per_second": 0.612, |
| "step": 2688 |
| }, |
| { |
| "epoch": 21.5, |
| "grad_norm": 1.7034225463867188, |
| "learning_rate": 4.616810498070709e-05, |
| "loss": 0.8461, |
| "step": 2752 |
| }, |
| { |
| "epoch": 21.5, |
| "eval_accuracy": 0.7483221476510067, |
| "eval_loss": 0.8213455677032471, |
| "eval_runtime": 8.1107, |
| "eval_samples_per_second": 36.742, |
| "eval_steps_per_second": 0.616, |
| "step": 2752 |
| }, |
| { |
| "epoch": 22.0, |
| "grad_norm": 1.6208754777908325, |
| "learning_rate": 4.058724504646835e-05, |
| "loss": 0.8179, |
| "step": 2816 |
| }, |
| { |
| "epoch": 22.0, |
| "eval_accuracy": 0.7416107382550335, |
| "eval_loss": 0.805648148059845, |
| "eval_runtime": 8.0669, |
| "eval_samples_per_second": 36.941, |
| "eval_steps_per_second": 0.62, |
| "step": 2816 |
| }, |
| { |
| "epoch": 22.5, |
| "grad_norm": 1.8578311204910278, |
| "learning_rate": 3.325697654887922e-05, |
| "loss": 0.8124, |
| "step": 2880 |
| }, |
| { |
| "epoch": 22.5, |
| "eval_accuracy": 0.7483221476510067, |
| "eval_loss": 0.8016577363014221, |
| "eval_runtime": 8.6546, |
| "eval_samples_per_second": 34.433, |
| "eval_steps_per_second": 0.578, |
| "step": 2880 |
| }, |
| { |
| "epoch": 23.0, |
| "grad_norm": 1.838831901550293, |
| "learning_rate": 2.5e-05, |
| "loss": 0.803, |
| "step": 2944 |
| }, |
| { |
| "epoch": 23.0, |
| "eval_accuracy": 0.7583892617449665, |
| "eval_loss": 0.7895504832267761, |
| "eval_runtime": 8.2476, |
| "eval_samples_per_second": 36.132, |
| "eval_steps_per_second": 0.606, |
| "step": 2944 |
| }, |
| { |
| "epoch": 23.5, |
| "grad_norm": 1.7440546751022339, |
| "learning_rate": 1.6743023451120788e-05, |
| "loss": 0.7979, |
| "step": 3008 |
| }, |
| { |
| "epoch": 23.5, |
| "eval_accuracy": 0.7583892617449665, |
| "eval_loss": 0.7843785881996155, |
| "eval_runtime": 8.6902, |
| "eval_samples_per_second": 34.292, |
| "eval_steps_per_second": 0.575, |
| "step": 3008 |
| }, |
| { |
| "epoch": 24.0, |
| "grad_norm": 1.838025450706482, |
| "learning_rate": 9.412754953531655e-06, |
| "loss": 0.7878, |
| "step": 3072 |
| }, |
| { |
| "epoch": 24.0, |
| "eval_accuracy": 0.7651006711409396, |
| "eval_loss": 0.7816659808158875, |
| "eval_runtime": 8.7178, |
| "eval_samples_per_second": 34.183, |
| "eval_steps_per_second": 0.574, |
| "step": 3072 |
| }, |
| { |
| "epoch": 24.5, |
| "grad_norm": 1.8526736497879028, |
| "learning_rate": 3.831895019292909e-06, |
| "loss": 0.7808, |
| "step": 3136 |
| }, |
| { |
| "epoch": 24.5, |
| "eval_accuracy": 0.761744966442953, |
| "eval_loss": 0.7814260721206665, |
| "eval_runtime": 8.7274, |
| "eval_samples_per_second": 34.145, |
| "eval_steps_per_second": 0.573, |
| "step": 3136 |
| }, |
| { |
| "epoch": 25.0, |
| "grad_norm": 1.6450481414794922, |
| "learning_rate": 6.268021954544068e-07, |
| "loss": 0.7924, |
| "step": 3200 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_accuracy": 0.761744966442953, |
| "eval_loss": 0.780437707901001, |
| "eval_runtime": 7.6749, |
| "eval_samples_per_second": 38.828, |
| "eval_steps_per_second": 0.651, |
| "step": 3200 |
| }, |
| { |
| "epoch": 25.5, |
| "grad_norm": 1.8034324645996094, |
| "learning_rate": 4.984280524733107e-05, |
| "loss": 0.8049, |
| "step": 3264 |
| }, |
| { |
| "epoch": 25.5, |
| "eval_accuracy": 0.761744966442953, |
| "eval_loss": 0.7750493288040161, |
| "eval_runtime": 8.6458, |
| "eval_samples_per_second": 34.468, |
| "eval_steps_per_second": 0.578, |
| "step": 3264 |
| }, |
| { |
| "epoch": 26.0, |
| "grad_norm": 1.6223326921463013, |
| "learning_rate": 4.75242216975605e-05, |
| "loss": 0.7581, |
| "step": 3328 |
| }, |
| { |
| "epoch": 26.0, |
| "eval_accuracy": 0.761744966442953, |
| "eval_loss": 0.7638412714004517, |
| "eval_runtime": 7.7052, |
| "eval_samples_per_second": 38.675, |
| "eval_steps_per_second": 0.649, |
| "step": 3328 |
| }, |
| { |
| "epoch": 26.5, |
| "grad_norm": 2.002840995788574, |
| "learning_rate": 4.267766952966369e-05, |
| "loss": 0.7651, |
| "step": 3392 |
| }, |
| { |
| "epoch": 26.5, |
| "eval_accuracy": 0.7684563758389261, |
| "eval_loss": 0.7540990710258484, |
| "eval_runtime": 7.6847, |
| "eval_samples_per_second": 38.779, |
| "eval_steps_per_second": 0.651, |
| "step": 3392 |
| }, |
| { |
| "epoch": 27.0, |
| "grad_norm": 1.705908179283142, |
| "learning_rate": 3.584709347793892e-05, |
| "loss": 0.7573, |
| "step": 3456 |
| }, |
| { |
| "epoch": 27.0, |
| "eval_accuracy": 0.7651006711409396, |
| "eval_loss": 0.7505062818527222, |
| "eval_runtime": 7.6264, |
| "eval_samples_per_second": 39.075, |
| "eval_steps_per_second": 0.656, |
| "step": 3456 |
| }, |
| { |
| "epoch": 27.5, |
| "grad_norm": 1.8837891817092896, |
| "learning_rate": 2.7799111902582686e-05, |
| "loss": 0.7524, |
| "step": 3520 |
| }, |
| { |
| "epoch": 27.5, |
| "eval_accuracy": 0.7651006711409396, |
| "eval_loss": 0.7437400221824646, |
| "eval_runtime": 7.6701, |
| "eval_samples_per_second": 38.852, |
| "eval_steps_per_second": 0.652, |
| "step": 3520 |
| }, |
| { |
| "epoch": 28.0, |
| "grad_norm": 1.8985482454299927, |
| "learning_rate": 1.9436976651092164e-05, |
| "loss": 0.7362, |
| "step": 3584 |
| }, |
| { |
| "epoch": 28.0, |
| "eval_accuracy": 0.7651006711409396, |
| "eval_loss": 0.7364336848258972, |
| "eval_runtime": 7.6555, |
| "eval_samples_per_second": 38.926, |
| "eval_steps_per_second": 0.653, |
| "step": 3584 |
| }, |
| { |
| "epoch": 28.5, |
| "grad_norm": 2.0026416778564453, |
| "learning_rate": 1.169919808711657e-05, |
| "loss": 0.7288, |
| "step": 3648 |
| }, |
| { |
| "epoch": 28.5, |
| "eval_accuracy": 0.7684563758389261, |
| "eval_loss": 0.7339101433753967, |
| "eval_runtime": 8.8394, |
| "eval_samples_per_second": 33.713, |
| "eval_steps_per_second": 0.566, |
| "step": 3648 |
| }, |
| { |
| "epoch": 29.0, |
| "grad_norm": 1.8134102821350098, |
| "learning_rate": 5.454212938299264e-06, |
| "loss": 0.7415, |
| "step": 3712 |
| }, |
| { |
| "epoch": 29.0, |
| "eval_accuracy": 0.7684563758389261, |
| "eval_loss": 0.7321557402610779, |
| "eval_runtime": 7.7574, |
| "eval_samples_per_second": 38.415, |
| "eval_steps_per_second": 0.645, |
| "step": 3712 |
| }, |
| { |
| "epoch": 29.5, |
| "grad_norm": 1.838896632194519, |
| "learning_rate": 1.4029167422908274e-06, |
| "loss": 0.7355, |
| "step": 3776 |
| }, |
| { |
| "epoch": 29.5, |
| "eval_accuracy": 0.7684563758389261, |
| "eval_loss": 0.7323412299156189, |
| "eval_runtime": 8.6786, |
| "eval_samples_per_second": 34.337, |
| "eval_steps_per_second": 0.576, |
| "step": 3776 |
| }, |
| { |
| "epoch": 30.0, |
| "grad_norm": 1.7982102632522583, |
| "learning_rate": 0.0, |
| "loss": 0.7283, |
| "step": 3840 |
| }, |
| { |
| "epoch": 30.0, |
| "eval_accuracy": 0.7684563758389261, |
| "eval_loss": 0.7321370244026184, |
| "eval_runtime": 8.9098, |
| "eval_samples_per_second": 33.446, |
| "eval_steps_per_second": 0.561, |
| "step": 3840 |
| }, |
| { |
| "epoch": 30.0, |
| "step": 3840, |
| "total_flos": 1.9045455502969405e+19, |
| "train_loss": 1.073897185921669, |
| "train_runtime": 8808.4478, |
| "train_samples_per_second": 27.9, |
| "train_steps_per_second": 0.436 |
| } |
| ], |
| "logging_steps": 64, |
| "max_steps": 3840, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 30, |
| "save_steps": 64, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.9045455502969405e+19, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
