{
  "best_metric": 1.0125610828399658,
  "best_model_checkpoint": "car_models_image_detection/checkpoint-32650",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 32650,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15313935681470137,
      "grad_norm": 1.452757477760315,
      "learning_rate": 1.4792944785276074e-05,
      "loss": 5.7276,
      "step": 500
    },
    {
      "epoch": 0.30627871362940273,
      "grad_norm": 1.6236835718154907,
      "learning_rate": 1.4562883435582821e-05,
      "loss": 5.577,
      "step": 1000
    },
    {
      "epoch": 0.45941807044410415,
      "grad_norm": 1.6465808153152466,
      "learning_rate": 1.433282208588957e-05,
      "loss": 5.415,
      "step": 1500
    },
    {
      "epoch": 0.6125574272588055,
      "grad_norm": 1.6408442258834839,
      "learning_rate": 1.410276073619632e-05,
      "loss": 5.2564,
      "step": 2000
    },
    {
      "epoch": 0.7656967840735069,
      "grad_norm": 2.1702845096588135,
      "learning_rate": 1.3872699386503068e-05,
      "loss": 5.097,
      "step": 2500
    },
    {
      "epoch": 0.9188361408882083,
      "grad_norm": 2.1283822059631348,
      "learning_rate": 1.3642638036809815e-05,
      "loss": 4.9628,
      "step": 3000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.27259150763221757,
      "eval_loss": 4.833654403686523,
      "eval_model_preparation_time": 0.0031,
      "eval_runtime": 685.3438,
      "eval_samples_per_second": 101.612,
      "eval_steps_per_second": 12.702,
      "step": 3265
    },
    {
      "epoch": 1.0719754977029097,
      "grad_norm": 2.1697254180908203,
      "learning_rate": 1.3412576687116564e-05,
      "loss": 4.8042,
      "step": 3500
    },
    {
      "epoch": 1.225114854517611,
      "grad_norm": 2.616407632827759,
      "learning_rate": 1.3182515337423314e-05,
      "loss": 4.6501,
      "step": 4000
    },
    {
      "epoch": 1.3782542113323124,
      "grad_norm": 2.611531972885132,
      "learning_rate": 1.2952453987730061e-05,
      "loss": 4.4964,
      "step": 4500
    },
    {
      "epoch": 1.5313935681470139,
      "grad_norm": 3.050095558166504,
      "learning_rate": 1.272239263803681e-05,
      "loss": 4.3653,
      "step": 5000
    },
    {
      "epoch": 1.6845329249617151,
      "grad_norm": 3.209902048110962,
      "learning_rate": 1.2492331288343558e-05,
      "loss": 4.2269,
      "step": 5500
    },
    {
      "epoch": 1.8376722817764164,
      "grad_norm": 3.8992629051208496,
      "learning_rate": 1.2262269938650307e-05,
      "loss": 4.0844,
      "step": 6000
    },
    {
      "epoch": 1.9908116385911179,
      "grad_norm": 3.587782382965088,
      "learning_rate": 1.2032208588957057e-05,
      "loss": 3.9577,
      "step": 6500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.432200347506426,
      "eval_loss": 3.934602975845337,
      "eval_model_preparation_time": 0.0031,
      "eval_runtime": 682.9317,
      "eval_samples_per_second": 101.971,
      "eval_steps_per_second": 12.747,
      "step": 6530
    },
    {
      "epoch": 2.1439509954058193,
      "grad_norm": 4.234837532043457,
      "learning_rate": 1.1802147239263804e-05,
      "loss": 3.7778,
      "step": 7000
    },
    {
      "epoch": 2.2970903522205206,
      "grad_norm": 3.8425912857055664,
      "learning_rate": 1.1572085889570552e-05,
      "loss": 3.6515,
      "step": 7500
    },
    {
      "epoch": 2.450229709035222,
      "grad_norm": 3.8768410682678223,
      "learning_rate": 1.1342024539877301e-05,
      "loss": 3.5171,
      "step": 8000
    },
    {
      "epoch": 2.6033690658499236,
      "grad_norm": 4.20833683013916,
      "learning_rate": 1.111196319018405e-05,
      "loss": 3.3911,
      "step": 8500
    },
    {
      "epoch": 2.756508422664625,
      "grad_norm": 4.394293785095215,
      "learning_rate": 1.0881901840490798e-05,
      "loss": 3.2882,
      "step": 9000
    },
    {
      "epoch": 2.909647779479326,
      "grad_norm": 4.391124725341797,
      "learning_rate": 1.0651840490797547e-05,
      "loss": 3.1777,
      "step": 9500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5524059794080903,
      "eval_loss": 3.125910520553589,
      "eval_model_preparation_time": 0.0031,
      "eval_runtime": 696.3871,
      "eval_samples_per_second": 100.0,
      "eval_steps_per_second": 12.5,
      "step": 9795
    },
    {
      "epoch": 3.0627871362940278,
      "grad_norm": 4.910506248474121,
      "learning_rate": 1.0421779141104295e-05,
      "loss": 3.0296,
      "step": 10000
    },
    {
      "epoch": 3.215926493108729,
      "grad_norm": 4.594521999359131,
      "learning_rate": 1.0191717791411042e-05,
      "loss": 2.8819,
      "step": 10500
    },
    {
      "epoch": 3.3690658499234303,
      "grad_norm": 4.765926361083984,
      "learning_rate": 9.961656441717793e-06,
      "loss": 2.7684,
      "step": 11000
    },
    {
      "epoch": 3.522205206738132,
      "grad_norm": 5.226629734039307,
      "learning_rate": 9.73159509202454e-06,
      "loss": 2.6588,
      "step": 11500
    },
    {
      "epoch": 3.6753445635528332,
      "grad_norm": 5.948498725891113,
      "learning_rate": 9.501533742331288e-06,
      "loss": 2.5607,
      "step": 12000
    },
    {
      "epoch": 3.8284839203675345,
      "grad_norm": 5.840036392211914,
      "learning_rate": 9.271472392638038e-06,
      "loss": 2.4664,
      "step": 12500
    },
    {
      "epoch": 3.9816232771822357,
      "grad_norm": 6.542360782623291,
      "learning_rate": 9.041411042944785e-06,
      "loss": 2.3674,
      "step": 13000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6481712833326153,
      "eval_loss": 2.4429006576538086,
      "eval_model_preparation_time": 0.0031,
      "eval_runtime": 700.3344,
      "eval_samples_per_second": 99.437,
      "eval_steps_per_second": 12.43,
      "step": 13060
    },
    {
      "epoch": 4.134762633996937,
      "grad_norm": 5.998870372772217,
      "learning_rate": 8.811349693251534e-06,
      "loss": 2.2206,
      "step": 13500
    },
    {
      "epoch": 4.287901990811639,
      "grad_norm": 5.736461639404297,
      "learning_rate": 8.581288343558282e-06,
      "loss": 2.1328,
      "step": 14000
    },
    {
      "epoch": 4.44104134762634,
      "grad_norm": 8.815576553344727,
      "learning_rate": 8.351226993865031e-06,
      "loss": 2.0378,
      "step": 14500
    },
    {
      "epoch": 4.594180704441041,
      "grad_norm": 6.433719635009766,
      "learning_rate": 8.121165644171779e-06,
      "loss": 2.0028,
      "step": 15000
    },
    {
      "epoch": 4.747320061255743,
      "grad_norm": 6.188543796539307,
      "learning_rate": 7.891104294478526e-06,
      "loss": 1.8992,
      "step": 15500
    },
    {
      "epoch": 4.900459418070444,
      "grad_norm": 4.191030979156494,
      "learning_rate": 7.661042944785277e-06,
      "loss": 1.8401,
      "step": 16000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7155042433119373,
      "eval_loss": 1.9256677627563477,
      "eval_model_preparation_time": 0.0031,
      "eval_runtime": 697.1043,
      "eval_samples_per_second": 99.898,
      "eval_steps_per_second": 12.487,
      "step": 16325
    },
    {
      "epoch": 5.053598774885145,
      "grad_norm": 6.046599388122559,
      "learning_rate": 7.430981595092025e-06,
      "loss": 1.749,
      "step": 16500
    },
    {
      "epoch": 5.206738131699847,
      "grad_norm": 7.202089309692383,
      "learning_rate": 7.200920245398773e-06,
      "loss": 1.6562,
      "step": 17000
    },
    {
      "epoch": 5.359877488514548,
      "grad_norm": 8.078461647033691,
      "learning_rate": 6.970858895705521e-06,
      "loss": 1.5944,
      "step": 17500
    },
    {
      "epoch": 5.51301684532925,
      "grad_norm": 8.001608848571777,
      "learning_rate": 6.74079754601227e-06,
      "loss": 1.517,
      "step": 18000
    },
    {
      "epoch": 5.666156202143951,
      "grad_norm": 7.847782135009766,
      "learning_rate": 6.510736196319019e-06,
      "loss": 1.4779,
      "step": 18500
    },
    {
      "epoch": 5.819295558958652,
      "grad_norm": 8.12964153289795,
      "learning_rate": 6.280674846625767e-06,
      "loss": 1.4133,
      "step": 19000
    },
    {
      "epoch": 5.972434915773354,
      "grad_norm": 5.552104473114014,
      "learning_rate": 6.0506134969325155e-06,
      "loss": 1.375,
      "step": 19500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7652464854463734,
      "eval_loss": 1.5537763833999634,
      "eval_model_preparation_time": 0.0031,
      "eval_runtime": 693.8343,
      "eval_samples_per_second": 100.368,
      "eval_steps_per_second": 12.546,
      "step": 19590
    },
    {
      "epoch": 6.1255742725880555,
      "grad_norm": 7.002144813537598,
      "learning_rate": 5.820552147239264e-06,
      "loss": 1.292,
      "step": 20000
    },
    {
      "epoch": 6.278713629402756,
      "grad_norm": 8.652531623840332,
      "learning_rate": 5.590490797546012e-06,
      "loss": 1.2451,
      "step": 20500
    },
    {
      "epoch": 6.431852986217458,
      "grad_norm": 6.782195568084717,
      "learning_rate": 5.360429447852761e-06,
      "loss": 1.199,
      "step": 21000
    },
    {
      "epoch": 6.584992343032159,
      "grad_norm": 7.46057653427124,
      "learning_rate": 5.13036809815951e-06,
      "loss": 1.1784,
      "step": 21500
    },
    {
      "epoch": 6.738131699846861,
      "grad_norm": 9.419745445251465,
      "learning_rate": 4.900306748466258e-06,
      "loss": 1.1378,
      "step": 22000
    },
    {
      "epoch": 6.891271056661562,
      "grad_norm": 10.585297584533691,
      "learning_rate": 4.670245398773006e-06,
      "loss": 1.1101,
      "step": 22500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8005858786025072,
      "eval_loss": 1.3030942678451538,
      "eval_model_preparation_time": 0.0031,
      "eval_runtime": 695.6789,
      "eval_samples_per_second": 100.102,
      "eval_steps_per_second": 12.513,
      "step": 22855
    },
    {
      "epoch": 7.044410413476263,
      "grad_norm": 6.227669715881348,
      "learning_rate": 4.440184049079755e-06,
      "loss": 1.0649,
      "step": 23000
    },
    {
      "epoch": 7.197549770290965,
      "grad_norm": 6.01344108581543,
      "learning_rate": 4.210122699386503e-06,
      "loss": 0.9895,
      "step": 23500
    },
    {
      "epoch": 7.3506891271056665,
      "grad_norm": 7.48536491394043,
      "learning_rate": 3.980061349693252e-06,
      "loss": 0.9931,
      "step": 24000
    },
    {
      "epoch": 7.503828483920367,
      "grad_norm": 4.995929718017578,
      "learning_rate": 3.75e-06,
      "loss": 0.9504,
      "step": 24500
    },
    {
      "epoch": 7.656967840735069,
      "grad_norm": 7.199100017547607,
      "learning_rate": 3.5199386503067485e-06,
      "loss": 0.9414,
      "step": 25000
    },
    {
      "epoch": 7.810107197549771,
      "grad_norm": 10.500569343566895,
      "learning_rate": 3.2898773006134974e-06,
      "loss": 0.9215,
      "step": 25500
    },
    {
      "epoch": 7.9632465543644715,
      "grad_norm": 12.734477043151855,
      "learning_rate": 3.0598159509202454e-06,
      "loss": 0.9171,
      "step": 26000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8239205043151108,
      "eval_loss": 1.1368967294692993,
      "eval_model_preparation_time": 0.0031,
      "eval_runtime": 700.1379,
      "eval_samples_per_second": 99.465,
      "eval_steps_per_second": 12.433,
      "step": 26120
    },
    {
      "epoch": 8.116385911179172,
      "grad_norm": 5.377818584442139,
      "learning_rate": 2.8297546012269938e-06,
      "loss": 0.877,
      "step": 26500
    },
    {
      "epoch": 8.269525267993874,
      "grad_norm": 8.250605583190918,
      "learning_rate": 2.599693251533742e-06,
      "loss": 0.8307,
      "step": 27000
    },
    {
      "epoch": 8.422664624808576,
      "grad_norm": 6.766539096832275,
      "learning_rate": 2.369631901840491e-06,
      "loss": 0.8152,
      "step": 27500
    },
    {
      "epoch": 8.575803981623277,
      "grad_norm": 8.64637565612793,
      "learning_rate": 2.1395705521472395e-06,
      "loss": 0.8146,
      "step": 28000
    },
    {
      "epoch": 8.728943338437979,
      "grad_norm": 7.545835494995117,
      "learning_rate": 1.9095092024539874e-06,
      "loss": 0.8026,
      "step": 28500
    },
    {
      "epoch": 8.88208269525268,
      "grad_norm": 7.024548053741455,
      "learning_rate": 1.6794478527607363e-06,
      "loss": 0.7875,
      "step": 29000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8368155774781373,
      "eval_loss": 1.044851541519165,
      "eval_model_preparation_time": 0.0031,
      "eval_runtime": 688.1062,
      "eval_samples_per_second": 101.204,
      "eval_steps_per_second": 12.651,
      "step": 29385
    },
    {
      "epoch": 9.03522205206738,
      "grad_norm": 6.825665473937988,
      "learning_rate": 1.4493865030674847e-06,
      "loss": 0.7609,
      "step": 29500
    },
    {
      "epoch": 9.188361408882082,
      "grad_norm": 8.946706771850586,
      "learning_rate": 1.2193251533742331e-06,
      "loss": 0.7448,
      "step": 30000
    },
    {
      "epoch": 9.341500765696784,
      "grad_norm": 5.960868835449219,
      "learning_rate": 9.892638036809815e-07,
      "loss": 0.7412,
      "step": 30500
    },
    {
      "epoch": 9.494640122511486,
      "grad_norm": 8.354199409484863,
      "learning_rate": 7.592024539877301e-07,
      "loss": 0.7391,
      "step": 31000
    },
    {
      "epoch": 9.647779479326188,
      "grad_norm": 7.826401233673096,
      "learning_rate": 5.291411042944785e-07,
      "loss": 0.7451,
      "step": 31500
    },
    {
      "epoch": 9.800918836140887,
      "grad_norm": 6.866321563720703,
      "learning_rate": 2.99079754601227e-07,
      "loss": 0.7328,
      "step": 32000
    },
    {
      "epoch": 9.95405819295559,
      "grad_norm": 8.034021377563477,
      "learning_rate": 6.901840490797547e-08,
      "loss": 0.7235,
      "step": 32500
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8410229899912406,
      "eval_loss": 1.0125610828399658,
      "eval_model_preparation_time": 0.0031,
      "eval_runtime": 702.1203,
      "eval_samples_per_second": 99.184,
      "eval_steps_per_second": 12.398,
      "step": 32650
    }
  ],
  "logging_steps": 500,
  "max_steps": 32650,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 8.117946756716657e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}