{
  "best_metric": 1.0466628074645996,
  "best_model_checkpoint": "./outputs/llava-mistral/RLAIF-V_Coocur-q0_50/checkpoint-200",
  "epoch": 1.391304347826087,
  "eval_steps": 50,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.034782608695652174,
      "grad_norm": 28.788983552687714,
      "learning_rate": 5e-07,
      "loss": 1.6631,
      "step": 5
    },
    {
      "epoch": 0.06956521739130435,
      "grad_norm": 14.801312030106265,
      "learning_rate": 1e-06,
      "loss": 1.5311,
      "step": 10
    },
    {
      "epoch": 0.10434782608695652,
      "grad_norm": 8.042597892857103,
      "learning_rate": 9.99648681635985e-07,
      "loss": 1.2909,
      "step": 15
    },
    {
      "epoch": 0.1391304347826087,
      "grad_norm": 5.465265266311333,
      "learning_rate": 9.985952202423114e-07,
      "loss": 1.2361,
      "step": 20
    },
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 5.471328096600488,
      "learning_rate": 9.96841096220313e-07,
      "loss": 1.2008,
      "step": 25
    },
    {
      "epoch": 0.20869565217391303,
      "grad_norm": 5.088536561695133,
      "learning_rate": 9.943887745939163e-07,
      "loss": 1.184,
      "step": 30
    },
    {
      "epoch": 0.24347826086956523,
      "grad_norm": 4.979056344341446,
      "learning_rate": 9.912417015456088e-07,
      "loss": 1.1743,
      "step": 35
    },
    {
      "epoch": 0.2782608695652174,
      "grad_norm": 5.098810591402351,
      "learning_rate": 9.874042995736093e-07,
      "loss": 1.1533,
      "step": 40
    },
    {
      "epoch": 0.3130434782608696,
      "grad_norm": 4.9260826588669575,
      "learning_rate": 9.828819612770495e-07,
      "loss": 1.15,
      "step": 45
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 5.009097422362774,
      "learning_rate": 9.77681041777897e-07,
      "loss": 1.1191,
      "step": 50
    },
    {
      "epoch": 0.34782608695652173,
      "eval_loss": 1.126644253730774,
      "eval_runtime": 71.6515,
      "eval_samples_per_second": 57.026,
      "eval_steps_per_second": 0.893,
      "step": 50
    },
    {
      "epoch": 0.3826086956521739,
      "grad_norm": 4.710201613179453,
      "learning_rate": 9.718088497902707e-07,
      "loss": 1.1278,
      "step": 55
    },
    {
      "epoch": 0.41739130434782606,
      "grad_norm": 4.656292887228332,
      "learning_rate": 9.652736373497e-07,
      "loss": 1.1259,
      "step": 60
    },
    {
      "epoch": 0.45217391304347826,
      "grad_norm": 4.89840217135744,
      "learning_rate": 9.580845882167572e-07,
      "loss": 1.1196,
      "step": 65
    },
    {
      "epoch": 0.48695652173913045,
      "grad_norm": 5.131971560708056,
      "learning_rate": 9.502518049713631e-07,
      "loss": 1.0982,
      "step": 70
    },
    {
      "epoch": 0.5217391304347826,
      "grad_norm": 5.025459861546626,
      "learning_rate": 9.417862948158997e-07,
      "loss": 1.1104,
      "step": 75
    },
    {
      "epoch": 0.5565217391304348,
      "grad_norm": 5.083697590841828,
      "learning_rate": 9.326999541070803e-07,
      "loss": 1.0856,
      "step": 80
    },
    {
      "epoch": 0.591304347826087,
      "grad_norm": 5.065321153827992,
      "learning_rate": 9.23005551638316e-07,
      "loss": 1.1057,
      "step": 85
    },
    {
      "epoch": 0.6260869565217392,
      "grad_norm": 4.793827429409211,
      "learning_rate": 9.127167106960681e-07,
      "loss": 1.0966,
      "step": 90
    },
    {
      "epoch": 0.6608695652173913,
      "grad_norm": 4.988686751743354,
      "learning_rate": 9.018478899154066e-07,
      "loss": 1.0984,
      "step": 95
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 5.137325195769105,
      "learning_rate": 8.904143629616732e-07,
      "loss": 1.0827,
      "step": 100
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 1.0803709030151367,
      "eval_runtime": 71.4447,
      "eval_samples_per_second": 57.191,
      "eval_steps_per_second": 0.896,
      "step": 100
    },
    {
      "epoch": 0.7304347826086957,
      "grad_norm": 5.24024786212869,
      "learning_rate": 8.784321970668053e-07,
      "loss": 1.0754,
      "step": 105
    },
    {
      "epoch": 0.7652173913043478,
      "grad_norm": 5.09088801959773,
      "learning_rate": 8.659182304504808e-07,
      "loss": 1.0796,
      "step": 110
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.068283621568658,
      "learning_rate": 8.528900486578158e-07,
      "loss": 1.0777,
      "step": 115
    },
    {
      "epoch": 0.8347826086956521,
      "grad_norm": 5.100518718656821,
      "learning_rate": 8.393659598468642e-07,
      "loss": 1.0909,
      "step": 120
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 5.090940478434361,
      "learning_rate": 8.253649690606494e-07,
      "loss": 1.0672,
      "step": 125
    },
    {
      "epoch": 0.9043478260869565,
      "grad_norm": 5.14647466195505,
      "learning_rate": 8.10906751519882e-07,
      "loss": 1.0672,
      "step": 130
    },
    {
      "epoch": 0.9391304347826087,
      "grad_norm": 4.93995490693938,
      "learning_rate": 7.960116249738937e-07,
      "loss": 1.056,
      "step": 135
    },
    {
      "epoch": 0.9739130434782609,
      "grad_norm": 5.093588987883723,
      "learning_rate": 7.807005211486444e-07,
      "loss": 1.0604,
      "step": 140
    },
    {
      "epoch": 1.008695652173913,
      "grad_norm": 5.489196503002481,
      "learning_rate": 7.649949563319227e-07,
      "loss": 1.0113,
      "step": 145
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 6.039150932058514,
      "learning_rate": 7.489170011370779e-07,
      "loss": 0.9347,
      "step": 150
    },
    {
      "epoch": 1.0434782608695652,
      "eval_loss": 1.0631003379821777,
      "eval_runtime": 71.1863,
      "eval_samples_per_second": 57.399,
      "eval_steps_per_second": 0.899,
      "step": 150
    },
    {
      "epoch": 1.0782608695652174,
      "grad_norm": 5.971802160671762,
      "learning_rate": 7.324892494877733e-07,
      "loss": 0.9277,
      "step": 155
    },
    {
      "epoch": 1.1130434782608696,
      "grad_norm": 5.391675039658058,
      "learning_rate": 7.15734786867344e-07,
      "loss": 0.9296,
      "step": 160
    },
    {
      "epoch": 1.1478260869565218,
      "grad_norm": 5.534633373232827,
      "learning_rate": 6.986771578773811e-07,
      "loss": 0.9147,
      "step": 165
    },
    {
      "epoch": 1.182608695652174,
      "grad_norm": 5.358757256728094,
      "learning_rate": 6.81340333151128e-07,
      "loss": 0.9163,
      "step": 170
    },
    {
      "epoch": 1.2173913043478262,
      "grad_norm": 5.543937956600803,
      "learning_rate": 6.637486756681842e-07,
      "loss": 0.915,
      "step": 175
    },
    {
      "epoch": 1.2521739130434781,
      "grad_norm": 5.443035617719796,
      "learning_rate": 6.459269065178591e-07,
      "loss": 0.9234,
      "step": 180
    },
    {
      "epoch": 1.2869565217391306,
      "grad_norm": 5.754731089663971,
      "learning_rate": 6.279000701592794e-07,
      "loss": 0.9208,
      "step": 185
    },
    {
      "epoch": 1.3217391304347825,
      "grad_norm": 5.688995590179024,
      "learning_rate": 6.096934992270767e-07,
      "loss": 0.9173,
      "step": 190
    },
    {
      "epoch": 1.3565217391304347,
      "grad_norm": 5.500423257037727,
      "learning_rate": 5.913327789321077e-07,
      "loss": 0.9141,
      "step": 195
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 5.827091513999703,
      "learning_rate": 5.728437111072375e-07,
      "loss": 0.9284,
      "step": 200
    },
    {
      "epoch": 1.391304347826087,
      "eval_loss": 1.0466628074645996,
      "eval_runtime": 71.2891,
      "eval_samples_per_second": 57.316,
      "eval_steps_per_second": 0.898,
      "step": 200
    }
  ],
  "logging_steps": 5,
  "max_steps": 429,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1179056703799296.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}