{ "best_metric": 4.781671524047852, "best_model_checkpoint": "car_models_image_detection/checkpoint-19590", "epoch": 6.0, "eval_steps": 500, "global_step": 19590, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.15313935681470137, "grad_norm": 1.3263182640075684, "learning_rate": 4.884851586489253e-06, "loss": 5.7627, "step": 500 }, { "epoch": 0.30627871362940273, "grad_norm": 1.3977476358413696, "learning_rate": 4.7569089048106455e-06, "loss": 5.7147, "step": 1000 }, { "epoch": 0.45941807044410415, "grad_norm": 1.7537243366241455, "learning_rate": 4.628966223132037e-06, "loss": 5.6711, "step": 1500 }, { "epoch": 0.6125574272588055, "grad_norm": 1.615222692489624, "learning_rate": 4.501023541453429e-06, "loss": 5.6233, "step": 2000 }, { "epoch": 0.7656967840735069, "grad_norm": 1.6928762197494507, "learning_rate": 4.3730808597748215e-06, "loss": 5.5761, "step": 2500 }, { "epoch": 0.9188361408882083, "grad_norm": 1.6801881790161133, "learning_rate": 4.245138178096213e-06, "loss": 5.5325, "step": 3000 }, { "epoch": 1.0, "eval_accuracy": 0.14117089561883428, "eval_loss": 5.504893779754639, "eval_model_preparation_time": 0.0051, "eval_runtime": 685.6796, "eval_samples_per_second": 101.562, "eval_steps_per_second": 12.695, "step": 3265 }, { "epoch": 1.0719754977029097, "grad_norm": 1.7748329639434814, "learning_rate": 4.117195496417605e-06, "loss": 5.4797, "step": 3500 }, { "epoch": 1.225114854517611, "grad_norm": 1.8438767194747925, "learning_rate": 3.9892528147389975e-06, "loss": 5.431, "step": 4000 }, { "epoch": 1.3782542113323124, "grad_norm": 1.7276830673217773, "learning_rate": 3.861310133060389e-06, "loss": 5.3855, "step": 4500 }, { "epoch": 1.5313935681470139, "grad_norm": 2.0461244583129883, "learning_rate": 3.7333674513817814e-06, "loss": 5.3484, "step": 5000 }, { "epoch": 1.6845329249617151, "grad_norm": 1.7156977653503418, "learning_rate": 3.605424769703173e-06, "loss": 5.3112, "step": 5500 }, { "epoch": 1.8376722817764164, "grad_norm": 1.7373435497283936, "learning_rate": 3.477482088024565e-06, "loss": 5.2742, "step": 6000 }, { "epoch": 1.9908116385911179, "grad_norm": 2.184741973876953, "learning_rate": 3.3495394063459574e-06, "loss": 5.2374, "step": 6500 }, { "epoch": 2.0, "eval_accuracy": 0.2263099699880814, "eval_loss": 5.24692964553833, "eval_model_preparation_time": 0.0051, "eval_runtime": 896.9055, "eval_samples_per_second": 77.644, "eval_steps_per_second": 9.706, "step": 6530 }, { "epoch": 2.1439509954058193, "grad_norm": 2.0060935020446777, "learning_rate": 3.221596724667349e-06, "loss": 5.188, "step": 7000 }, { "epoch": 2.2970903522205206, "grad_norm": 2.189157247543335, "learning_rate": 3.093654042988741e-06, "loss": 5.149, "step": 7500 }, { "epoch": 2.450229709035222, "grad_norm": 1.9108648300170898, "learning_rate": 2.9657113613101333e-06, "loss": 5.1341, "step": 8000 }, { "epoch": 2.6033690658499236, "grad_norm": 2.189399003982544, "learning_rate": 2.837768679631525e-06, "loss": 5.0976, "step": 8500 }, { "epoch": 2.756508422664625, "grad_norm": 1.9839409589767456, "learning_rate": 2.709825997952917e-06, "loss": 5.0666, "step": 9000 }, { "epoch": 2.909647779479326, "grad_norm": 2.16378116607666, "learning_rate": 2.5818833162743093e-06, "loss": 5.0364, "step": 9500 }, { "epoch": 3.0, "eval_accuracy": 0.28031706371429804, "eval_loss": 5.047238349914551, "eval_model_preparation_time": 0.0051, "eval_runtime": 693.5793, "eval_samples_per_second": 100.405, "eval_steps_per_second": 12.551, "step": 9795 }, { "epoch": 3.0627871362940278, "grad_norm": 2.0013692378997803, "learning_rate": 2.453940634595701e-06, "loss": 4.9956, "step": 10000 }, { "epoch": 3.215926493108729, "grad_norm": 2.1350488662719727, "learning_rate": 2.3259979529170936e-06, "loss": 4.9729, "step": 10500 }, { "epoch": 3.3690658499234303, "grad_norm": 2.2013564109802246, "learning_rate": 2.1980552712384853e-06, "loss": 4.9543, "step": 11000 }, { "epoch": 3.522205206738132, "grad_norm": 2.276151657104492, "learning_rate": 2.0701125895598775e-06, "loss": 4.9241, "step": 11500 }, { "epoch": 3.6753445635528332, "grad_norm": 2.143611192703247, "learning_rate": 1.9421699078812696e-06, "loss": 4.9079, "step": 12000 }, { "epoch": 3.8284839203675345, "grad_norm": 2.168256998062134, "learning_rate": 1.8142272262026613e-06, "loss": 4.8768, "step": 12500 }, { "epoch": 3.9816232771822357, "grad_norm": 2.3494791984558105, "learning_rate": 1.6862845445240533e-06, "loss": 4.8585, "step": 13000 }, { "epoch": 4.0, "eval_accuracy": 0.31762374531512516, "eval_loss": 4.901946067810059, "eval_model_preparation_time": 0.0051, "eval_runtime": 701.2958, "eval_samples_per_second": 99.3, "eval_steps_per_second": 12.413, "step": 13060 }, { "epoch": 4.134762633996937, "grad_norm": 2.179281234741211, "learning_rate": 1.5583418628454452e-06, "loss": 4.8325, "step": 13500 }, { "epoch": 4.287901990811639, "grad_norm": 2.776350975036621, "learning_rate": 1.4303991811668375e-06, "loss": 4.8248, "step": 14000 }, { "epoch": 4.44104134762634, "grad_norm": 2.138899087905884, "learning_rate": 1.3024564994882295e-06, "loss": 4.8024, "step": 14500 }, { "epoch": 4.594180704441041, "grad_norm": 2.3762903213500977, "learning_rate": 1.1745138178096214e-06, "loss": 4.7856, "step": 15000 }, { "epoch": 4.747320061255743, "grad_norm": 2.3697316646575928, "learning_rate": 1.0465711361310133e-06, "loss": 4.7752, "step": 15500 }, { "epoch": 4.900459418070444, "grad_norm": 2.36728835105896, "learning_rate": 9.186284544524055e-07, "loss": 4.7718, "step": 16000 }, { "epoch": 5.0, "eval_accuracy": 0.3355303780927354, "eval_loss": 4.811797618865967, "eval_model_preparation_time": 0.0051, "eval_runtime": 709.5406, "eval_samples_per_second": 98.147, "eval_steps_per_second": 12.269, "step": 16325 }, { "epoch": 5.053598774885145, "grad_norm": 2.806079864501953, "learning_rate": 7.906857727737974e-07, "loss": 4.7554, "step": 16500 }, { "epoch": 5.206738131699847, "grad_norm": 2.5930612087249756, "learning_rate": 6.627430910951894e-07, "loss": 4.7433, "step": 17000 }, { "epoch": 5.359877488514548, "grad_norm": 2.619678258895874, "learning_rate": 5.348004094165814e-07, "loss": 4.738, "step": 17500 }, { "epoch": 5.51301684532925, "grad_norm": 2.2594709396362305, "learning_rate": 4.068577277379734e-07, "loss": 4.7254, "step": 18000 }, { "epoch": 5.666156202143951, "grad_norm": 2.825565814971924, "learning_rate": 2.7891504605936546e-07, "loss": 4.718, "step": 18500 }, { "epoch": 5.819295558958652, "grad_norm": 2.3414411544799805, "learning_rate": 1.5097236438075744e-07, "loss": 4.7316, "step": 19000 }, { "epoch": 5.972434915773354, "grad_norm": 2.9150233268737793, "learning_rate": 2.3029682702149437e-08, "loss": 4.7157, "step": 19500 }, { "epoch": 6.0, "eval_accuracy": 0.34041269978029554, "eval_loss": 4.781671524047852, "eval_model_preparation_time": 0.0051, "eval_runtime": 712.1452, "eval_samples_per_second": 97.788, "eval_steps_per_second": 12.224, "step": 19590 } ], "logging_steps": 500, "max_steps": 19590, "num_input_tokens_seen": 0, "num_train_epochs": 6, "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": 4.8707680540299936e+19, "train_batch_size": 32, "trial_name": null, "trial_params": null }