{
  "best_metric": 1.039311408996582,
  "best_model_checkpoint": "./outputs/llava-mistral/RLAIF-V_Coocur-q0_50/checkpoint-400",
  "epoch": 2.9843478260869567,
  "eval_steps": 50,
  "global_step": 429,
  "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
    },
    {
      "epoch": 1.4260869565217391,
      "grad_norm": 5.802245310068559,
      "learning_rate": 5.542522779487071e-07,
      "loss": 0.9138,
      "step": 205
    },
    {
      "epoch": 1.4608695652173913,
      "grad_norm": 5.61384072841023,
      "learning_rate": 5.355846055040448e-07,
      "loss": 0.8997,
      "step": 210
    },
    {
      "epoch": 1.4956521739130435,
      "grad_norm": 5.904579252940991,
      "learning_rate": 5.168669269578232e-07,
      "loss": 0.8884,
      "step": 215
    },
    {
      "epoch": 1.5304347826086957,
      "grad_norm": 5.271432415804347,
      "learning_rate": 4.981255457668624e-07,
      "loss": 0.8992,
      "step": 220
    },
    {
      "epoch": 1.5652173913043477,
      "grad_norm": 5.899020033427604,
      "learning_rate": 4.793867986966802e-07,
      "loss": 0.8925,
      "step": 225
    },
    {
      "epoch": 1.6,
      "grad_norm": 5.350779976715096,
      "learning_rate": 4.606770188111338e-07,
      "loss": 0.8914,
      "step": 230
    },
    {
      "epoch": 1.634782608695652,
      "grad_norm": 5.512269864038317,
      "learning_rate": 4.420224984672653e-07,
      "loss": 0.8912,
      "step": 235
    },
    {
      "epoch": 1.6695652173913045,
      "grad_norm": 5.678688235761292,
      "learning_rate": 4.2344945236734963e-07,
      "loss": 0.9089,
      "step": 240
    },
    {
      "epoch": 1.7043478260869565,
      "grad_norm": 5.627490372225429,
      "learning_rate": 4.049839807200688e-07,
      "loss": 0.9176,
      "step": 245
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 5.641266523304515,
      "learning_rate": 3.866520325625825e-07,
      "loss": 0.9103,
      "step": 250
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 1.03276789188385,
      "eval_runtime": 71.4756,
      "eval_samples_per_second": 57.166,
      "eval_steps_per_second": 0.895,
      "step": 250
    },
    {
      "epoch": 1.7739130434782608,
      "grad_norm": 5.527194295602311,
      "learning_rate": 3.684793692950344e-07,
      "loss": 0.9032,
      "step": 255
    },
    {
      "epoch": 1.808695652173913,
      "grad_norm": 5.767218475736747,
      "learning_rate": 3.504915284787405e-07,
      "loss": 0.8957,
      "step": 260
    },
    {
      "epoch": 1.8434782608695652,
      "grad_norm": 5.498699782720146,
      "learning_rate": 3.327137879489312e-07,
      "loss": 0.8956,
      "step": 265
    },
    {
      "epoch": 1.8782608695652174,
      "grad_norm": 5.5888005763392,
      "learning_rate": 3.1517113029248233e-07,
      "loss": 0.8961,
      "step": 270
    },
    {
      "epoch": 1.9130434782608696,
      "grad_norm": 5.6876581687221766,
      "learning_rate": 2.9788820774054697e-07,
      "loss": 0.8908,
      "step": 275
    },
    {
      "epoch": 1.9478260869565216,
      "grad_norm": 6.233330329395755,
      "learning_rate": 2.8088930752543063e-07,
      "loss": 0.8993,
      "step": 280
    },
    {
      "epoch": 1.982608695652174,
      "grad_norm": 5.614794208303309,
      "learning_rate": 2.641983177503876e-07,
      "loss": 0.8856,
      "step": 285
    },
    {
      "epoch": 2.017391304347826,
      "grad_norm": 6.378827429420443,
      "learning_rate": 2.4783869382030424e-07,
      "loss": 0.8606,
      "step": 290
    },
    {
      "epoch": 2.0521739130434784,
      "grad_norm": 6.12687179373102,
      "learning_rate": 2.3183342548044065e-07,
      "loss": 0.8044,
      "step": 295
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 5.8322466278221805,
      "learning_rate": 2.1620500450955221e-07,
      "loss": 0.8121,
      "step": 300
    },
    {
      "epoch": 2.0869565217391304,
      "eval_loss": 1.0422885417938232,
      "eval_runtime": 71.4365,
      "eval_samples_per_second": 57.198,
      "eval_steps_per_second": 0.896,
      "step": 300
    },
    {
      "epoch": 2.121739130434783,
      "grad_norm": 6.6256496639714495,
      "learning_rate": 2.0097539311278898e-07,
      "loss": 0.8008,
      "step": 305
    },
    {
      "epoch": 2.1565217391304348,
      "grad_norm": 5.9761400221530225,
      "learning_rate": 1.8616599305879333e-07,
      "loss": 0.8045,
      "step": 310
    },
    {
      "epoch": 2.1913043478260867,
      "grad_norm": 5.9592246987241895,
      "learning_rate": 1.7179761560436097e-07,
      "loss": 0.7988,
      "step": 315
    },
    {
      "epoch": 2.226086956521739,
      "grad_norm": 5.657294423926494,
      "learning_rate": 1.5789045224893538e-07,
      "loss": 0.8104,
      "step": 320
    },
    {
      "epoch": 2.260869565217391,
      "grad_norm": 5.953621888664232,
      "learning_rate": 1.444640463600293e-07,
      "loss": 0.8145,
      "step": 325
    },
    {
      "epoch": 2.2956521739130435,
      "grad_norm": 5.7958284830752795,
      "learning_rate": 1.3153726570944828e-07,
      "loss": 0.8036,
      "step": 330
    },
    {
      "epoch": 2.3304347826086955,
      "grad_norm": 6.277490040902002,
      "learning_rate": 1.1912827595891312e-07,
      "loss": 0.8045,
      "step": 335
    },
    {
      "epoch": 2.365217391304348,
      "grad_norm": 6.071597440507695,
      "learning_rate": 1.0725451513233674e-07,
      "loss": 0.8015,
      "step": 340
    },
    {
      "epoch": 2.4,
      "grad_norm": 6.117435343058851,
      "learning_rate": 9.593266911063253e-08,
      "loss": 0.791,
      "step": 345
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 6.209598217279743,
      "learning_rate": 8.517864818348803e-08,
      "loss": 0.7971,
      "step": 350
    },
    {
      "epoch": 2.4347826086956523,
      "eval_loss": 1.039114236831665,
      "eval_runtime": 71.581,
      "eval_samples_per_second": 57.082,
      "eval_steps_per_second": 0.894,
      "step": 350
    },
    {
      "epoch": 2.4695652173913043,
      "grad_norm": 6.032280132363507,
      "learning_rate": 7.500756469105818e-08,
      "loss": 0.7921,
      "step": 355
    },
    {
      "epoch": 2.5043478260869563,
      "grad_norm": 6.142136260190064,
      "learning_rate": 6.543371178699442e-08,
      "loss": 0.7977,
      "step": 360
    },
    {
      "epoch": 2.5391304347826087,
      "grad_norm": 6.121634316544134,
      "learning_rate": 5.647054335265489e-08,
      "loss": 0.8018,
      "step": 365
    },
    {
      "epoch": 2.573913043478261,
      "grad_norm": 6.258148278180407,
      "learning_rate": 4.813065509072278e-08,
      "loss": 0.8162,
      "step": 370
    },
    {
      "epoch": 2.608695652173913,
      "grad_norm": 6.37201864445784,
      "learning_rate": 4.0425766824798814e-08,
      "loss": 0.7968,
      "step": 375
    },
    {
      "epoch": 2.643478260869565,
      "grad_norm": 6.053989181642519,
      "learning_rate": 3.3366706029845096e-08,
      "loss": 0.7995,
      "step": 380
    },
    {
      "epoch": 2.6782608695652175,
      "grad_norm": 6.115299861345008,
      "learning_rate": 2.696339261662156e-08,
      "loss": 0.8093,
      "step": 385
    },
    {
      "epoch": 2.7130434782608694,
      "grad_norm": 6.081706575040738,
      "learning_rate": 2.122482499149869e-08,
      "loss": 0.7993,
      "step": 390
    },
    {
      "epoch": 2.747826086956522,
      "grad_norm": 5.938881075296142,
      "learning_rate": 1.6159067411235737e-08,
      "loss": 0.7798,
      "step": 395
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 5.619310965697272,
      "learning_rate": 1.177323865049512e-08,
      "loss": 0.8019,
      "step": 400
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 1.039311408996582,
      "eval_runtime": 71.5502,
      "eval_samples_per_second": 57.107,
      "eval_steps_per_second": 0.894,
      "step": 400
    },
    {
      "epoch": 2.8173913043478263,
      "grad_norm": 6.071621720496628,
      "learning_rate": 8.073501998017152e-09,
      "loss": 0.8007,
      "step": 405
    },
    {
      "epoch": 2.8521739130434782,
      "grad_norm": 6.290477888624567,
      "learning_rate": 5.065056595513984e-09,
      "loss": 0.7815,
      "step": 410
    },
    {
      "epoch": 2.8869565217391306,
      "grad_norm": 5.900577794332532,
      "learning_rate": 2.752130131453756e-09,
      "loss": 0.7902,
      "step": 415
    },
    {
      "epoch": 2.9217391304347826,
      "grad_norm": 6.116774590773252,
      "learning_rate": 1.137972900002171e-09,
      "loss": 0.8089,
      "step": 420
    },
    {
      "epoch": 2.9565217391304346,
      "grad_norm": 6.333281708927235,
      "learning_rate": 2.2485323347054552e-10,
      "loss": 0.8092,
      "step": 425
    },
    {
      "epoch": 2.9843478260869567,
      "step": 429,
      "total_flos": 2529344492666880.0,
      "train_loss": 0.9544596938820152,
      "train_runtime": 6643.7341,
      "train_samples_per_second": 16.604,
      "train_steps_per_second": 0.065
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2529344492666880.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}