| { | |
| "best_metric": 1.572628378868103, | |
| "best_model_checkpoint": "cat_breed_image_detection/checkpoint-10164", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 10164, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19677292404565133, | |
| "grad_norm": 8.132095336914062, | |
| "learning_rate": 8.599564959462131e-07, | |
| "loss": 1.6107, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39354584809130266, | |
| "grad_norm": 6.708084583282471, | |
| "learning_rate": 8.154637136642278e-07, | |
| "loss": 1.5888, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5903187721369539, | |
| "grad_norm": 7.669999122619629, | |
| "learning_rate": 7.709709313822424e-07, | |
| "loss": 1.5734, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7870916961826053, | |
| "grad_norm": 8.473257064819336, | |
| "learning_rate": 7.26478149100257e-07, | |
| "loss": 1.5921, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9838646202282566, | |
| "grad_norm": 7.519883155822754, | |
| "learning_rate": 6.819853668182716e-07, | |
| "loss": 1.5814, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.583766976675524, | |
| "eval_loss": 1.6136096715927124, | |
| "eval_runtime": 1206.4218, | |
| "eval_samples_per_second": 89.839, | |
| "eval_steps_per_second": 2.807, | |
| "step": 2541 | |
| }, | |
| { | |
| "epoch": 1.1806375442739079, | |
| "grad_norm": 6.776809215545654, | |
| "learning_rate": 6.374925845362863e-07, | |
| "loss": 1.5698, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.3774104683195592, | |
| "grad_norm": 5.961708068847656, | |
| "learning_rate": 5.929998022543009e-07, | |
| "loss": 1.5564, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.5741833923652107, | |
| "grad_norm": 7.470553398132324, | |
| "learning_rate": 5.485070199723155e-07, | |
| "loss": 1.5616, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.770956316410862, | |
| "grad_norm": 6.852038383483887, | |
| "learning_rate": 5.040142376903302e-07, | |
| "loss": 1.5534, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.9677292404565132, | |
| "grad_norm": 8.469923973083496, | |
| "learning_rate": 4.5952145540834485e-07, | |
| "loss": 1.5467, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5894412459403602, | |
| "eval_loss": 1.5893911123275757, | |
| "eval_runtime": 1210.5447, | |
| "eval_samples_per_second": 89.533, | |
| "eval_steps_per_second": 2.798, | |
| "step": 5082 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "grad_norm": 6.9930195808410645, | |
| "learning_rate": 4.150286731263595e-07, | |
| "loss": 1.5455, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.3612750885478158, | |
| "grad_norm": 8.173332214355469, | |
| "learning_rate": 3.705358908443741e-07, | |
| "loss": 1.5393, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.558048012593467, | |
| "grad_norm": 5.5481767654418945, | |
| "learning_rate": 3.2604310856238877e-07, | |
| "loss": 1.5412, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.7548209366391183, | |
| "grad_norm": 6.221193313598633, | |
| "learning_rate": 2.815503262804034e-07, | |
| "loss": 1.531, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.9515938606847696, | |
| "grad_norm": 6.431015968322754, | |
| "learning_rate": 2.37057543998418e-07, | |
| "loss": 1.5297, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5934916592855034, | |
| "eval_loss": 1.5773507356643677, | |
| "eval_runtime": 1196.383, | |
| "eval_samples_per_second": 90.593, | |
| "eval_steps_per_second": 2.831, | |
| "step": 7623 | |
| }, | |
| { | |
| "epoch": 3.1483667847304213, | |
| "grad_norm": 10.443075180053711, | |
| "learning_rate": 1.9256476171643265e-07, | |
| "loss": 1.5376, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.3451397087760726, | |
| "grad_norm": 6.658329010009766, | |
| "learning_rate": 1.4807197943444728e-07, | |
| "loss": 1.5207, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.541912632821724, | |
| "grad_norm": 6.124561786651611, | |
| "learning_rate": 1.0357919715246192e-07, | |
| "loss": 1.5225, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.738685556867375, | |
| "grad_norm": 7.456569194793701, | |
| "learning_rate": 5.9086414870476566e-08, | |
| "loss": 1.5198, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.9354584809130264, | |
| "grad_norm": 6.399494647979736, | |
| "learning_rate": 1.4593632588491198e-08, | |
| "loss": 1.5279, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5949955713020372, | |
| "eval_loss": 1.572628378868103, | |
| "eval_runtime": 1198.3774, | |
| "eval_samples_per_second": 90.442, | |
| "eval_steps_per_second": 2.826, | |
| "step": 10164 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10164, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.0414127158862545e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |