{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.995910949568378,
  "eval_steps": 400,
  "global_step": 137,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03634711494775102,
      "grad_norm": 35.987906218257066,
      "learning_rate": 2.857142857142857e-07,
      "logits/chosen": -11.111563682556152,
      "logits/rejected": -11.135547637939453,
      "logps/chosen": -0.662230372428894,
      "logps/rejected": -0.6628919839859009,
      "loss": 5.0952,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -6.6223039627075195,
      "rewards/margins": 0.006614901125431061,
      "rewards/rejected": -6.6289191246032715,
      "step": 5
    },
    {
      "epoch": 0.07269422989550205,
      "grad_norm": 28.899244168425763,
      "learning_rate": 5.714285714285714e-07,
      "logits/chosen": -10.829585075378418,
      "logits/rejected": -10.654484748840332,
      "logps/chosen": -0.636147677898407,
      "logps/rejected": -0.6354170441627502,
      "loss": 5.0407,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -6.361477375030518,
      "rewards/margins": -0.007306712679564953,
      "rewards/rejected": -6.354170799255371,
      "step": 10
    },
    {
      "epoch": 0.10904134484325306,
      "grad_norm": 34.054577665859114,
      "learning_rate": 7.998695344323425e-07,
      "logits/chosen": -10.697937965393066,
      "logits/rejected": -10.68556022644043,
      "logps/chosen": -0.756206214427948,
      "logps/rejected": -0.7363730669021606,
      "loss": 5.0934,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": -7.562062740325928,
      "rewards/margins": -0.1983323097229004,
      "rewards/rejected": -7.363729953765869,
      "step": 15
    },
    {
      "epoch": 0.1453884597910041,
      "grad_norm": 31.063536076780956,
      "learning_rate": 7.953121695121394e-07,
      "logits/chosen": -10.083427429199219,
      "logits/rejected": -9.985143661499023,
      "logps/chosen": -0.6148006319999695,
      "logps/rejected": -0.6226206421852112,
      "loss": 5.024,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": -6.148005485534668,
      "rewards/margins": 0.07820063084363937,
      "rewards/rejected": -6.226205825805664,
      "step": 20
    },
    {
      "epoch": 0.18173557473875512,
      "grad_norm": 37.55152432622566,
      "learning_rate": 7.843163833184991e-07,
      "logits/chosen": -9.981136322021484,
      "logits/rejected": -9.983463287353516,
      "logps/chosen": -0.5722960233688354,
      "logps/rejected": -0.5717057585716248,
      "loss": 4.961,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -5.722960472106934,
      "rewards/margins": -0.005903184413909912,
      "rewards/rejected": -5.717057228088379,
      "step": 25
    },
    {
      "epoch": 0.21808268968650613,
      "grad_norm": 29.53598266606176,
      "learning_rate": 7.670612634414511e-07,
      "logits/chosen": -9.952229499816895,
      "logits/rejected": -9.804238319396973,
      "logps/chosen": -0.5058998465538025,
      "logps/rejected": -0.5249067544937134,
      "loss": 4.8596,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -5.058998107910156,
      "rewards/margins": 0.19006863236427307,
      "rewards/rejected": -5.249067306518555,
      "step": 30
    },
    {
      "epoch": 0.25442980463425713,
      "grad_norm": 30.836147418383828,
      "learning_rate": 7.438278427948805e-07,
      "logits/chosen": -9.969837188720703,
      "logits/rejected": -9.884472846984863,
      "logps/chosen": -0.535486102104187,
      "logps/rejected": -0.5739948153495789,
      "loss": 4.7794,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -5.354860782623291,
      "rewards/margins": 0.3850872814655304,
      "rewards/rejected": -5.739948272705078,
      "step": 35
    },
    {
      "epoch": 0.2907769195820082,
      "grad_norm": 35.784369740177794,
      "learning_rate": 7.149945224533862e-07,
      "logits/chosen": -9.953130722045898,
      "logits/rejected": -9.866512298583984,
      "logps/chosen": -0.6358538866043091,
      "logps/rejected": -0.65082848072052,
      "loss": 4.8879,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -6.358539581298828,
      "rewards/margins": 0.14974501729011536,
      "rewards/rejected": -6.508284091949463,
      "step": 40
    },
    {
      "epoch": 0.3271240345297592,
      "grad_norm": 30.738414397371837,
      "learning_rate": 6.810309086608129e-07,
      "logits/chosen": -10.294210433959961,
      "logits/rejected": -10.260662078857422,
      "logps/chosen": -0.6186165809631348,
      "logps/rejected": -0.642761766910553,
      "loss": 4.817,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -6.186165809631348,
      "rewards/margins": 0.24145162105560303,
      "rewards/rejected": -6.427617073059082,
      "step": 45
    },
    {
      "epoch": 0.36347114947751025,
      "grad_norm": 46.12276326749997,
      "learning_rate": 6.424901643866552e-07,
      "logits/chosen": -10.330435752868652,
      "logits/rejected": -10.143304824829102,
      "logps/chosen": -0.6831095814704895,
      "logps/rejected": -0.7464486956596375,
      "loss": 4.6885,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -6.8310956954956055,
      "rewards/margins": 0.6333915591239929,
      "rewards/rejected": -7.464486598968506,
      "step": 50
    },
    {
      "epoch": 0.39981826442526125,
      "grad_norm": 43.40349458380626,
      "learning_rate": 6e-07,
      "logits/chosen": -11.197371482849121,
      "logits/rejected": -11.12946605682373,
      "logps/chosen": -0.5718105435371399,
      "logps/rejected": -0.6250694394111633,
      "loss": 4.6953,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -5.718105792999268,
      "rewards/margins": 0.5325883626937866,
      "rewards/rejected": -6.250694274902344,
      "step": 55
    },
    {
      "epoch": 0.43616537937301225,
      "grad_norm": 62.775385851545494,
      "learning_rate": 5.542524497952543e-07,
      "logits/chosen": -10.749040603637695,
      "logits/rejected": -10.889803886413574,
      "logps/chosen": -0.6364445686340332,
      "logps/rejected": -0.7200480699539185,
      "loss": 4.6271,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -6.364445686340332,
      "rewards/margins": 0.8360347747802734,
      "rewards/rejected": -7.200479984283447,
      "step": 60
    },
    {
      "epoch": 0.4725124943207633,
      "grad_norm": 45.98517277447237,
      "learning_rate": 5.059926008786647e-07,
      "logits/chosen": -11.873740196228027,
      "logits/rejected": -11.824403762817383,
      "logps/chosen": -0.7099884748458862,
      "logps/rejected": -0.7525407075881958,
      "loss": 4.638,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -7.0998854637146,
      "rewards/margins": 0.4255215525627136,
      "rewards/rejected": -7.525406837463379,
      "step": 65
    },
    {
      "epoch": 0.5088596092685143,
      "grad_norm": 47.30579548391997,
      "learning_rate": 4.5600645798745166e-07,
      "logits/chosen": -12.138381004333496,
      "logits/rejected": -11.983818054199219,
      "logps/chosen": -0.6762068867683411,
      "logps/rejected": -0.7381676435470581,
      "loss": 4.703,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -6.762068748474121,
      "rewards/margins": 0.619607150554657,
      "rewards/rejected": -7.38167667388916,
      "step": 70
    },
    {
      "epoch": 0.5452067242162654,
      "grad_norm": 69.5932279945552,
      "learning_rate": 4.051081418863895e-07,
      "logits/chosen": -12.561568260192871,
      "logits/rejected": -12.478489875793457,
      "logps/chosen": -0.6981841325759888,
      "logps/rejected": -0.7842732667922974,
      "loss": 4.5323,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -6.98184061050415,
      "rewards/margins": 0.8608924150466919,
      "rewards/rejected": -7.842732906341553,
      "step": 75
    },
    {
      "epoch": 0.5815538391640164,
      "grad_norm": 127.53517276446773,
      "learning_rate": 3.541266298406398e-07,
      "logits/chosen": -12.076669692993164,
      "logits/rejected": -12.142119407653809,
      "logps/chosen": -0.7188806533813477,
      "logps/rejected": -0.7781528234481812,
      "loss": 4.6379,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -7.188807010650635,
      "rewards/margins": 0.5927222371101379,
      "rewards/rejected": -7.781529426574707,
      "step": 80
    },
    {
      "epoch": 0.6179009541117674,
      "grad_norm": 61.22401801870787,
      "learning_rate": 3.0389225412181565e-07,
      "logits/chosen": -13.094499588012695,
      "logits/rejected": -13.159561157226562,
      "logps/chosen": -0.7800406813621521,
      "logps/rejected": -0.8725376129150391,
      "loss": 4.4548,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -7.800407409667969,
      "rewards/margins": 0.9249688386917114,
      "rewards/rejected": -8.725375175476074,
      "step": 85
    },
    {
      "epoch": 0.6542480690595184,
      "grad_norm": 66.37522679776315,
      "learning_rate": 2.5522317844515273e-07,
      "logits/chosen": -13.562166213989258,
      "logits/rejected": -13.433174133300781,
      "logps/chosen": -0.7740485072135925,
      "logps/rejected": -0.8862950205802917,
      "loss": 4.3623,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -7.740484714508057,
      "rewards/margins": 1.1224651336669922,
      "rewards/rejected": -8.862950325012207,
      "step": 90
    },
    {
      "epoch": 0.6905951840072694,
      "grad_norm": 72.9710577397891,
      "learning_rate": 2.0891207259509476e-07,
      "logits/chosen": -14.113569259643555,
      "logits/rejected": -13.75220012664795,
      "logps/chosen": -0.7652403116226196,
      "logps/rejected": -0.8429893255233765,
      "loss": 4.3962,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -7.652403354644775,
      "rewards/margins": 0.7774893045425415,
      "rewards/rejected": -8.429892539978027,
      "step": 95
    },
    {
      "epoch": 0.7269422989550205,
      "grad_norm": 60.63378631631187,
      "learning_rate": 1.6571320226872206e-07,
      "logits/chosen": -14.494039535522461,
      "logits/rejected": -14.628949165344238,
      "logps/chosen": -0.7868660688400269,
      "logps/rejected": -0.9212242960929871,
      "loss": 4.1386,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -7.868659973144531,
      "rewards/margins": 1.3435838222503662,
      "rewards/rejected": -9.212244033813477,
      "step": 100
    },
    {
      "epoch": 0.7632894139027715,
      "grad_norm": 190.4622656460041,
      "learning_rate": 1.2633014440382787e-07,
      "logits/chosen": -15.230148315429688,
      "logits/rejected": -14.929168701171875,
      "logps/chosen": -0.8822343945503235,
      "logps/rejected": -0.9908519983291626,
      "loss": 4.2736,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -8.822343826293945,
      "rewards/margins": 1.0861766338348389,
      "rewards/rejected": -9.908520698547363,
      "step": 105
    },
    {
      "epoch": 0.7996365288505225,
      "grad_norm": 101.38847574260835,
      "learning_rate": 9.14043280712228e-08,
      "logits/chosen": -15.496833801269531,
      "logits/rejected": -15.636553764343262,
      "logps/chosen": -0.8570655584335327,
      "logps/rejected": -0.9334220886230469,
      "loss": 4.2832,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -8.570656776428223,
      "rewards/margins": 0.7635642886161804,
      "rewards/rejected": -9.334220886230469,
      "step": 110
    },
    {
      "epoch": 0.8359836437982735,
      "grad_norm": 74.21593181905395,
      "learning_rate": 6.150458756494239e-08,
      "logits/chosen": -15.494606018066406,
      "logits/rejected": -15.465006828308105,
      "logps/chosen": -0.9054273366928101,
      "logps/rejected": -1.0216588973999023,
      "loss": 4.1426,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -9.05427360534668,
      "rewards/margins": 1.1623156070709229,
      "rewards/rejected": -10.216588973999023,
      "step": 115
    },
    {
      "epoch": 0.8723307587460245,
      "grad_norm": 65.9311757106038,
      "learning_rate": 3.711789783843522e-08,
      "logits/chosen": -15.544687271118164,
      "logits/rejected": -15.537959098815918,
      "logps/chosen": -0.8991163969039917,
      "logps/rejected": -0.9694231748580933,
      "loss": 4.2329,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -8.991164207458496,
      "rewards/margins": 0.7030680179595947,
      "rewards/rejected": -9.694231033325195,
      "step": 120
    },
    {
      "epoch": 0.9086778736937755,
      "grad_norm": 130.83461356205703,
      "learning_rate": 1.8641443178027784e-08,
      "logits/chosen": -15.450345039367676,
      "logits/rejected": -15.695470809936523,
      "logps/chosen": -0.8895367383956909,
      "logps/rejected": -1.0279381275177002,
      "loss": 4.1031,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -8.895367622375488,
      "rewards/margins": 1.3840134143829346,
      "rewards/rejected": -10.279379844665527,
      "step": 125
    },
    {
      "epoch": 0.9450249886415266,
      "grad_norm": 82.30298232430259,
      "learning_rate": 6.376148290617145e-09,
      "logits/chosen": -16.2006778717041,
      "logits/rejected": -16.138973236083984,
      "logps/chosen": -0.9445899724960327,
      "logps/rejected": -1.0684407949447632,
      "loss": 4.0298,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -9.445899963378906,
      "rewards/margins": 1.2385092973709106,
      "rewards/rejected": -10.684408187866211,
      "step": 130
    },
    {
      "epoch": 0.9813721035892776,
      "grad_norm": 88.42391053771068,
      "learning_rate": 5.217771643080127e-10,
      "logits/chosen": -16.4208927154541,
      "logits/rejected": -16.113683700561523,
      "logps/chosen": -0.8756990432739258,
      "logps/rejected": -0.9942779541015625,
      "loss": 4.1263,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -8.756990432739258,
      "rewards/margins": 1.1857887506484985,
      "rewards/rejected": -9.942779541015625,
      "step": 135
    },
    {
      "epoch": 0.995910949568378,
      "step": 137,
      "total_flos": 0.0,
      "train_loss": 4.581099252631194,
      "train_runtime": 3146.1495,
      "train_samples_per_second": 5.596,
      "train_steps_per_second": 0.044
    }
  ],
  "logging_steps": 5,
  "max_steps": 137,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}