{
  "best_metric": 1.0907771587371826,
  "best_model_checkpoint": "./outputs/llava-mistral/RLAIF-V_Coocur-q0_25/checkpoint-200",
  "epoch": 3.0,
  "eval_steps": 50,
  "global_step": 216,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06944444444444445,
      "grad_norm": 28.445383103621488,
      "learning_rate": 5e-07,
      "loss": 1.692,
      "step": 5
    },
    {
      "epoch": 0.1388888888888889,
      "grad_norm": 15.214710637322117,
      "learning_rate": 1e-06,
      "loss": 1.5428,
      "step": 10
    },
    {
      "epoch": 0.20833333333333334,
      "grad_norm": 7.725515360957365,
      "learning_rate": 9.985471028179154e-07,
      "loss": 1.3232,
      "step": 15
    },
    {
      "epoch": 0.2777777777777778,
      "grad_norm": 5.771045580757361,
      "learning_rate": 9.94196854912548e-07,
      "loss": 1.2575,
      "step": 20
    },
    {
      "epoch": 0.3472222222222222,
      "grad_norm": 5.8072030206153045,
      "learning_rate": 9.869745381355905e-07,
      "loss": 1.2163,
      "step": 25
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 5.2740216973392755,
      "learning_rate": 9.769221256218162e-07,
      "loss": 1.1937,
      "step": 30
    },
    {
      "epoch": 0.4861111111111111,
      "grad_norm": 5.030592874663904,
      "learning_rate": 9.64098037858483e-07,
      "loss": 1.1473,
      "step": 35
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 4.82797980028618,
      "learning_rate": 9.485768031694871e-07,
      "loss": 1.1529,
      "step": 40
    },
    {
      "epoch": 0.625,
      "grad_norm": 5.01413477732448,
      "learning_rate": 9.304486245873971e-07,
      "loss": 1.1598,
      "step": 45
    },
    {
      "epoch": 0.6944444444444444,
      "grad_norm": 5.005624161351076,
      "learning_rate": 9.098188556305262e-07,
      "loss": 1.1343,
      "step": 50
    },
    {
      "epoch": 0.6944444444444444,
      "eval_loss": 1.1338481903076172,
      "eval_runtime": 35.4083,
      "eval_samples_per_second": 57.698,
      "eval_steps_per_second": 0.904,
      "step": 50
    },
    {
      "epoch": 0.7638888888888888,
      "grad_norm": 5.121349492890328,
      "learning_rate": 8.868073880316123e-07,
      "loss": 1.1465,
      "step": 55
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 4.967730285097845,
      "learning_rate": 8.615479549763755e-07,
      "loss": 1.1205,
      "step": 60
    },
    {
      "epoch": 0.9027777777777778,
      "grad_norm": 4.888543325436437,
      "learning_rate": 8.341873539012443e-07,
      "loss": 1.1101,
      "step": 65
    },
    {
      "epoch": 0.9722222222222222,
      "grad_norm": 5.136971847854665,
      "learning_rate": 8.048845933670271e-07,
      "loss": 1.1137,
      "step": 70
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 5.65186903458913,
      "learning_rate": 7.738099689665539e-07,
      "loss": 1.0344,
      "step": 75
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 5.436495818544836,
      "learning_rate": 7.41144073636728e-07,
      "loss": 0.9903,
      "step": 80
    },
    {
      "epoch": 1.1805555555555556,
      "grad_norm": 5.1217292921898485,
      "learning_rate": 7.070767481266492e-07,
      "loss": 0.976,
      "step": 85
    },
    {
      "epoch": 1.25,
      "grad_norm": 5.550754536917889,
      "learning_rate": 6.718059777212565e-07,
      "loss": 0.9731,
      "step": 90
    },
    {
      "epoch": 1.3194444444444444,
      "grad_norm": 5.3291194115834095,
      "learning_rate": 6.355367416322778e-07,
      "loss": 0.9687,
      "step": 95
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 5.187329472162837,
      "learning_rate": 5.984798217433531e-07,
      "loss": 0.9522,
      "step": 100
    },
    {
      "epoch": 1.3888888888888888,
      "eval_loss": 1.0978167057037354,
      "eval_runtime": 35.0302,
      "eval_samples_per_second": 58.321,
      "eval_steps_per_second": 0.913,
      "step": 100
    },
    {
      "epoch": 1.4583333333333333,
      "grad_norm": 5.429628497748933,
      "learning_rate": 5.608505776324157e-07,
      "loss": 0.9579,
      "step": 105
    },
    {
      "epoch": 1.5277777777777777,
      "grad_norm": 5.186690517651686,
      "learning_rate": 5.228676949903973e-07,
      "loss": 0.9593,
      "step": 110
    },
    {
      "epoch": 1.5972222222222223,
      "grad_norm": 5.242848282685733,
      "learning_rate": 4.847519147099294e-07,
      "loss": 0.952,
      "step": 115
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 5.57786673198586,
      "learning_rate": 4.46724750030062e-07,
      "loss": 0.9444,
      "step": 120
    },
    {
      "epoch": 1.7361111111111112,
      "grad_norm": 5.398693342092869,
      "learning_rate": 4.0900719919241935e-07,
      "loss": 0.9505,
      "step": 125
    },
    {
      "epoch": 1.8055555555555556,
      "grad_norm": 5.747268409628116,
      "learning_rate": 3.7181846109031e-07,
      "loss": 0.9547,
      "step": 130
    },
    {
      "epoch": 1.875,
      "grad_norm": 5.567481347529886,
      "learning_rate": 3.353746613749093e-07,
      "loss": 0.9463,
      "step": 135
    },
    {
      "epoch": 1.9444444444444444,
      "grad_norm": 5.400936083455457,
      "learning_rate": 2.9988759642186093e-07,
      "loss": 0.943,
      "step": 140
    },
    {
      "epoch": 2.013888888888889,
      "grad_norm": 6.366675658176883,
      "learning_rate": 2.655635024578483e-07,
      "loss": 0.9177,
      "step": 145
    },
    {
      "epoch": 2.0833333333333335,
      "grad_norm": 5.430229049864778,
      "learning_rate": 2.3260185700046292e-07,
      "loss": 0.8681,
      "step": 150
    },
    {
      "epoch": 2.0833333333333335,
      "eval_loss": 1.084751844406128,
      "eval_runtime": 35.1219,
      "eval_samples_per_second": 58.169,
      "eval_steps_per_second": 0.911,
      "step": 150
    },
    {
      "epoch": 2.1527777777777777,
      "grad_norm": 5.491519565547442,
      "learning_rate": 2.0119421957691218e-07,
      "loss": 0.842,
      "step": 155
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 5.753980047210634,
      "learning_rate": 1.7152311845883094e-07,
      "loss": 0.8516,
      "step": 160
    },
    {
      "epoch": 2.2916666666666665,
      "grad_norm": 5.773460825469727,
      "learning_rate": 1.4376098988303404e-07,
      "loss": 0.8471,
      "step": 165
    },
    {
      "epoch": 2.361111111111111,
      "grad_norm": 5.780876141809209,
      "learning_rate": 1.1806917592302761e-07,
      "loss": 0.8699,
      "step": 170
    },
    {
      "epoch": 2.4305555555555554,
      "grad_norm": 5.458710876272155,
      "learning_rate": 9.459698683523204e-08,
      "loss": 0.8268,
      "step": 175
    },
    {
      "epoch": 2.5,
      "grad_norm": 5.709014238644116,
      "learning_rate": 7.348083332917926e-08,
      "loss": 0.8705,
      "step": 180
    },
    {
      "epoch": 2.5694444444444446,
      "grad_norm": 5.718392280522503,
      "learning_rate": 5.484343380457124e-08,
      "loss": 0.8466,
      "step": 185
    },
    {
      "epoch": 2.638888888888889,
      "grad_norm": 5.770791652413458,
      "learning_rate": 3.879310116241041e-08,
      "loss": 0.8482,
      "step": 190
    },
    {
      "epoch": 2.7083333333333335,
      "grad_norm": 5.555636942265803,
      "learning_rate": 2.5423113334966218e-08,
      "loss": 0.8456,
      "step": 195
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 5.6835384408720735,
      "learning_rate": 1.4811171192794624e-08,
      "loss": 0.8498,
      "step": 200
    },
    {
      "epoch": 2.7777777777777777,
      "eval_loss": 1.0907771587371826,
      "eval_runtime": 35.1999,
      "eval_samples_per_second": 58.04,
      "eval_steps_per_second": 0.909,
      "step": 200
    },
    {
      "epoch": 2.8472222222222223,
      "grad_norm": 5.604525039241507,
      "learning_rate": 7.018946979234997e-09,
      "loss": 0.8546,
      "step": 205
    },
    {
      "epoch": 2.9166666666666665,
      "grad_norm": 5.794165801523241,
      "learning_rate": 2.0917258966953734e-09,
      "loss": 0.8421,
      "step": 210
    },
    {
      "epoch": 2.986111111111111,
      "grad_norm": 5.761167232022704,
      "learning_rate": 5.814292768108187e-11,
      "loss": 0.8491,
      "step": 215
    },
    {
      "epoch": 3.0,
      "step": 216,
      "total_flos": 1273399955423232.0,
      "train_loss": 1.0138301187091403,
      "train_runtime": 3161.6084,
      "train_samples_per_second": 17.445,
      "train_steps_per_second": 0.068
    }
  ],
  "logging_steps": 5,
  "max_steps": 216,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1273399955423232.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}