|
{ |
|
"best_metric": 2.90333890914917, |
|
"best_model_checkpoint": "car_brands_image_detection/checkpoint-13060", |
|
"epoch": 4.0, |
|
"eval_steps": 500, |
|
"global_step": 13060, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.15, |
|
"grad_norm": 6.046504497528076, |
|
"learning_rate": 9.654112221368178e-07, |
|
"loss": 2.9162, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"grad_norm": 4.840131759643555, |
|
"learning_rate": 9.269792467332821e-07, |
|
"loss": 2.9167, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"grad_norm": 5.4137043952941895, |
|
"learning_rate": 8.885472713297463e-07, |
|
"loss": 2.9148, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"grad_norm": 4.974403381347656, |
|
"learning_rate": 8.501152959262105e-07, |
|
"loss": 2.8946, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"grad_norm": 5.067456245422363, |
|
"learning_rate": 8.116833205226749e-07, |
|
"loss": 2.8872, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"grad_norm": 6.574715614318848, |
|
"learning_rate": 7.732513451191392e-07, |
|
"loss": 2.8665, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.6150576544752222, |
|
"eval_loss": 2.9889485836029053, |
|
"eval_runtime": 698.2446, |
|
"eval_samples_per_second": 99.734, |
|
"eval_steps_per_second": 12.467, |
|
"step": 3265 |
|
}, |
|
{ |
|
"epoch": 1.07, |
|
"grad_norm": 6.3792195320129395, |
|
"learning_rate": 7.348193697156033e-07, |
|
"loss": 2.8564, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 1.23, |
|
"grad_norm": 9.3410062789917, |
|
"learning_rate": 6.963873943120676e-07, |
|
"loss": 2.8661, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.38, |
|
"grad_norm": 7.090749263763428, |
|
"learning_rate": 6.579554189085319e-07, |
|
"loss": 2.8411, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"grad_norm": 6.649363994598389, |
|
"learning_rate": 6.19523443504996e-07, |
|
"loss": 2.8276, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"grad_norm": 5.353057384490967, |
|
"learning_rate": 5.810914681014604e-07, |
|
"loss": 2.8096, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"grad_norm": 6.126298904418945, |
|
"learning_rate": 5.426594926979247e-07, |
|
"loss": 2.8139, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"grad_norm": 5.249317646026611, |
|
"learning_rate": 5.042275172943889e-07, |
|
"loss": 2.8157, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.6201266531684831, |
|
"eval_loss": 2.9412875175476074, |
|
"eval_runtime": 695.3694, |
|
"eval_samples_per_second": 100.147, |
|
"eval_steps_per_second": 12.519, |
|
"step": 6530 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"grad_norm": 7.586893558502197, |
|
"learning_rate": 4.6579554189085316e-07, |
|
"loss": 2.813, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 2.3, |
|
"grad_norm": 5.9368109703063965, |
|
"learning_rate": 4.2736356648731743e-07, |
|
"loss": 2.8071, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 2.45, |
|
"grad_norm": 5.697512626647949, |
|
"learning_rate": 3.889315910837817e-07, |
|
"loss": 2.8052, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 2.6, |
|
"grad_norm": 5.205171585083008, |
|
"learning_rate": 3.5049961568024597e-07, |
|
"loss": 2.7838, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 2.76, |
|
"grad_norm": 4.8574066162109375, |
|
"learning_rate": 3.1206764027671024e-07, |
|
"loss": 2.7709, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 2.91, |
|
"grad_norm": 6.57193660736084, |
|
"learning_rate": 2.7363566487317445e-07, |
|
"loss": 2.7698, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.6226396128606098, |
|
"eval_loss": 2.913849353790283, |
|
"eval_runtime": 686.1905, |
|
"eval_samples_per_second": 101.486, |
|
"eval_steps_per_second": 12.686, |
|
"step": 9795 |
|
}, |
|
{ |
|
"epoch": 3.06, |
|
"grad_norm": 7.7096171379089355, |
|
"learning_rate": 2.3520368946963872e-07, |
|
"loss": 2.7689, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 3.22, |
|
"grad_norm": 6.814896583557129, |
|
"learning_rate": 1.9677171406610299e-07, |
|
"loss": 2.7725, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 3.37, |
|
"grad_norm": 5.8311028480529785, |
|
"learning_rate": 1.5833973866256725e-07, |
|
"loss": 2.7611, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 3.52, |
|
"grad_norm": 5.887447834014893, |
|
"learning_rate": 1.1990776325903152e-07, |
|
"loss": 2.7716, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 3.68, |
|
"grad_norm": 8.307939529418945, |
|
"learning_rate": 8.147578785549578e-08, |
|
"loss": 2.7494, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 3.83, |
|
"grad_norm": 5.97913122177124, |
|
"learning_rate": 4.304381245196003e-08, |
|
"loss": 2.7573, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 3.98, |
|
"grad_norm": 6.363702774047852, |
|
"learning_rate": 4.611837048424288e-09, |
|
"loss": 2.7635, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.6236878760464682, |
|
"eval_loss": 2.90333890914917, |
|
"eval_runtime": 690.5493, |
|
"eval_samples_per_second": 100.846, |
|
"eval_steps_per_second": 12.606, |
|
"step": 13060 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 13060, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 4, |
|
"save_steps": 500, |
|
"total_flos": 3.247178702686662e+19, |
|
"train_batch_size": 32, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|