| { | |
| "best_metric": 1.5611639022827148, | |
| "best_model_checkpoint": "67_cat_breeds_image_detection/checkpoint-25328", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 25328, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.821979586992643e-06, | |
| "loss": 3.1114, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 9.624179128095577e-06, | |
| "loss": 2.6215, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 9.426378669198514e-06, | |
| "loss": 2.4034, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 9.228578210301448e-06, | |
| "loss": 2.2396, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 9.030777751404383e-06, | |
| "loss": 2.1317, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 8.83297729250732e-06, | |
| "loss": 2.0335, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5173761946133797, | |
| "eval_loss": 1.9850196838378906, | |
| "eval_runtime": 296.7547, | |
| "eval_samples_per_second": 85.33, | |
| "eval_steps_per_second": 10.669, | |
| "step": 3166 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 8.635176833610254e-06, | |
| "loss": 1.9677, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 8.437376374713189e-06, | |
| "loss": 1.9292, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 8.239575915816125e-06, | |
| "loss": 1.8358, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 8.04177545691906e-06, | |
| "loss": 1.8098, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 7.843974998021997e-06, | |
| "loss": 1.7845, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 7.646174539124931e-06, | |
| "loss": 1.7439, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5424926940999921, | |
| "eval_loss": 1.7416865825653076, | |
| "eval_runtime": 305.3072, | |
| "eval_samples_per_second": 82.939, | |
| "eval_steps_per_second": 10.37, | |
| "step": 6332 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 7.448374080227867e-06, | |
| "loss": 1.7391, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 7.250573621330802e-06, | |
| "loss": 1.678, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 7.052773162433737e-06, | |
| "loss": 1.6614, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 6.8549727035366735e-06, | |
| "loss": 1.6771, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 6.657172244639608e-06, | |
| "loss": 1.6238, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 6.459371785742543e-06, | |
| "loss": 1.6347, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5577758470894874, | |
| "eval_loss": 1.64887535572052, | |
| "eval_runtime": 301.4384, | |
| "eval_samples_per_second": 84.004, | |
| "eval_steps_per_second": 10.503, | |
| "step": 9498 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 6.261571326845479e-06, | |
| "loss": 1.6022, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 6.063770867948414e-06, | |
| "loss": 1.5793, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 5.865970409051349e-06, | |
| "loss": 1.5794, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 5.668169950154285e-06, | |
| "loss": 1.5531, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 5.47036949125722e-06, | |
| "loss": 1.564, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 5.2725690323601555e-06, | |
| "loss": 1.5424, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 5.074768573463091e-06, | |
| "loss": 1.5426, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5635415843930179, | |
| "eval_loss": 1.6023435592651367, | |
| "eval_runtime": 296.7308, | |
| "eval_samples_per_second": 85.337, | |
| "eval_steps_per_second": 10.67, | |
| "step": 12664 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 4.876968114566027e-06, | |
| "loss": 1.5209, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 4.679167655668961e-06, | |
| "loss": 1.518, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 4.481367196771897e-06, | |
| "loss": 1.4888, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 4.2835667378748325e-06, | |
| "loss": 1.499, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 4.085766278977767e-06, | |
| "loss": 1.4888, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 3.887965820080703e-06, | |
| "loss": 1.4907, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5614485427691336, | |
| "eval_loss": 1.5871776342391968, | |
| "eval_runtime": 298.3387, | |
| "eval_samples_per_second": 84.877, | |
| "eval_steps_per_second": 10.612, | |
| "step": 15830 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 3.690165361183638e-06, | |
| "loss": 1.4685, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 3.4923649022865734e-06, | |
| "loss": 1.4495, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 3.294564443389509e-06, | |
| "loss": 1.4432, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 3.096763984492444e-06, | |
| "loss": 1.449, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 2.8989635255953797e-06, | |
| "loss": 1.4445, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 2.7011630666983153e-06, | |
| "loss": 1.4592, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5697022352104889, | |
| "eval_loss": 1.5659719705581665, | |
| "eval_runtime": 299.3375, | |
| "eval_samples_per_second": 84.593, | |
| "eval_steps_per_second": 10.577, | |
| "step": 18996 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 2.50336260780125e-06, | |
| "loss": 1.4347, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 2.3055621489041856e-06, | |
| "loss": 1.4233, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 2.107761690007121e-06, | |
| "loss": 1.4259, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 1.9099612311100563e-06, | |
| "loss": 1.4131, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 1.7121607722129916e-06, | |
| "loss": 1.3964, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 1.5143603133159272e-06, | |
| "loss": 1.4152, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 1.3165598544188623e-06, | |
| "loss": 1.4004, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5668193665587237, | |
| "eval_loss": 1.565866231918335, | |
| "eval_runtime": 298.5036, | |
| "eval_samples_per_second": 84.83, | |
| "eval_steps_per_second": 10.606, | |
| "step": 22162 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 1.1187593955217977e-06, | |
| "loss": 1.4024, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 9.20958936624733e-07, | |
| "loss": 1.4075, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 7.231584777276685e-07, | |
| "loss": 1.3715, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 5.253580188306037e-07, | |
| "loss": 1.3919, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 3.275575599335391e-07, | |
| "loss": 1.384, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 7.9, | |
| "learning_rate": 1.2975710103647441e-07, | |
| "loss": 1.384, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5691493562909723, | |
| "eval_loss": 1.5611639022827148, | |
| "eval_runtime": 298.9869, | |
| "eval_samples_per_second": 84.693, | |
| "eval_steps_per_second": 10.589, | |
| "step": 25328 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 25328, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "total_flos": 6.282678931597911e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |