{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7562189054726368,
  "eval_steps": 500,
  "global_step": 1900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03980099502487562,
      "grad_norm": 2.466599941253662,
      "learning_rate": 4.980474401576887e-07,
      "logits/chosen": -1.0382839441299438,
      "logits/rejected": -1.054585576057434,
      "logps/chosen": -180.16372680664062,
      "logps/rejected": -179.63958740234375,
      "loss": 0.7169,
      "rewards/accuracies": 0.508593738079071,
      "rewards/chosen": -0.278283029794693,
      "rewards/margins": 0.017729664221405983,
      "rewards/rejected": -0.2960126996040344,
      "step": 100
    },
    {
      "epoch": 0.07960199004975124,
      "grad_norm": 1.6285502910614014,
      "learning_rate": 4.922202605502572e-07,
      "logits/chosen": -1.0206574201583862,
      "logits/rejected": -1.0379180908203125,
      "logps/chosen": -180.13119506835938,
      "logps/rejected": -180.29315185546875,
      "loss": 0.7103,
      "rewards/accuracies": 0.5193750262260437,
      "rewards/chosen": -0.2392772138118744,
      "rewards/margins": 0.0195913203060627,
      "rewards/rejected": -0.2588685154914856,
      "step": 200
    },
    {
      "epoch": 0.11940298507462686,
      "grad_norm": 2.7344954013824463,
      "learning_rate": 4.82609484512869e-07,
      "logits/chosen": -1.0241189002990723,
      "logits/rejected": -1.0348433256149292,
      "logps/chosen": -178.86837768554688,
      "logps/rejected": -179.53598022460938,
      "loss": 0.7077,
      "rewards/accuracies": 0.5171874761581421,
      "rewards/chosen": -0.21391521394252777,
      "rewards/margins": 0.015260601416230202,
      "rewards/rejected": -0.22917583584785461,
      "step": 300
    },
    {
      "epoch": 0.15920398009950248,
      "grad_norm": 1.7553608417510986,
      "learning_rate": 4.6936523696827614e-07,
      "logits/chosen": -1.0034430027008057,
      "logits/rejected": -1.0131057500839233,
      "logps/chosen": -180.70912170410156,
      "logps/rejected": -181.44435119628906,
      "loss": 0.7064,
      "rewards/accuracies": 0.5106250047683716,
      "rewards/chosen": -0.18936260044574738,
      "rewards/margins": 0.013838082551956177,
      "rewards/rejected": -0.20320068299770355,
      "step": 400
    },
    {
      "epoch": 0.19900497512437812,
      "grad_norm": 1.4984805583953857,
      "learning_rate": 4.5269439940365644e-07,
      "logits/chosen": -1.0015250444412231,
      "logits/rejected": -1.0229138135910034,
      "logps/chosen": -178.01638793945312,
      "logps/rejected": -177.6285400390625,
      "loss": 0.7093,
      "rewards/accuracies": 0.4985937476158142,
      "rewards/chosen": -0.1724223494529724,
      "rewards/margins": 0.005522388964891434,
      "rewards/rejected": -0.17794474959373474,
      "step": 500
    },
    {
      "epoch": 0.23880597014925373,
      "grad_norm": 1.539502501487732,
      "learning_rate": 4.328573782827409e-07,
      "logits/chosen": -0.9833173155784607,
      "logits/rejected": -1.0065593719482422,
      "logps/chosen": -176.92718505859375,
      "logps/rejected": -174.7305145263672,
      "loss": 0.7017,
      "rewards/accuracies": 0.507031261920929,
      "rewards/chosen": -0.14655204117298126,
      "rewards/margins": 0.01735287345945835,
      "rewards/rejected": -0.16390492022037506,
      "step": 600
    },
    {
      "epoch": 0.27860696517412936,
      "grad_norm": 1.7529277801513672,
      "learning_rate": 4.1016403737218373e-07,
      "logits/chosen": -0.9706727862358093,
      "logits/rejected": -0.997480034828186,
      "logps/chosen": -179.5781707763672,
      "logps/rejected": -174.51856994628906,
      "loss": 0.7024,
      "rewards/accuracies": 0.5096874833106995,
      "rewards/chosen": -0.1383100152015686,
      "rewards/margins": 0.014245204627513885,
      "rewards/rejected": -0.1525552123785019,
      "step": 700
    },
    {
      "epoch": 0.31840796019900497,
      "grad_norm": 1.58861243724823,
      "learning_rate": 3.849688575211836e-07,
      "logits/chosen": -0.9722832441329956,
      "logits/rejected": -0.9961209893226624,
      "logps/chosen": -182.63351440429688,
      "logps/rejected": -178.73550415039062,
      "loss": 0.6995,
      "rewards/accuracies": 0.5214062333106995,
      "rewards/chosen": -0.12984366714954376,
      "rewards/margins": 0.020543336868286133,
      "rewards/rejected": -0.1503870040178299,
      "step": 800
    },
    {
      "epoch": 0.3582089552238806,
      "grad_norm": 1.5765061378479004,
      "learning_rate": 3.576653995009154e-07,
      "logits/chosen": -0.9867467284202576,
      "logits/rejected": -1.0024300813674927,
      "logps/chosen": -175.5293426513672,
      "logps/rejected": -175.31863403320312,
      "loss": 0.6951,
      "rewards/accuracies": 0.5267187356948853,
      "rewards/chosen": -0.11608893424272537,
      "rewards/margins": 0.024803265929222107,
      "rewards/rejected": -0.14089219272136688,
      "step": 900
    },
    {
      "epoch": 0.39800995024875624,
      "grad_norm": 1.4879308938980103,
      "learning_rate": 3.286801563968721e-07,
      "logits/chosen": -0.9625688195228577,
      "logits/rejected": -0.9898955821990967,
      "logps/chosen": -178.4788818359375,
      "logps/rejected": -175.9660186767578,
      "loss": 0.6984,
      "rewards/accuracies": 0.5181249976158142,
      "rewards/chosen": -0.11111991107463837,
      "rewards/margins": 0.020023101940751076,
      "rewards/rejected": -0.1311430186033249,
      "step": 1000
    },
    {
      "epoch": 0.43781094527363185,
      "grad_norm": 1.4386117458343506,
      "learning_rate": 2.9846589158269034e-07,
      "logits/chosen": -0.9723007082939148,
      "logits/rejected": -0.9996936321258545,
      "logps/chosen": -181.09291076660156,
      "logps/rejected": -176.69898986816406,
      "loss": 0.7013,
      "rewards/accuracies": 0.5106250047683716,
      "rewards/chosen": -0.11242911964654922,
      "rewards/margins": 0.0125286765396595,
      "rewards/rejected": -0.12495779991149902,
      "step": 1100
    },
    {
      "epoch": 0.47761194029850745,
      "grad_norm": 2.057227373123169,
      "learning_rate": 2.674945663394993e-07,
      "logits/chosen": -0.9660943746566772,
      "logits/rejected": -0.9956172108650208,
      "logps/chosen": -177.69273376464844,
      "logps/rejected": -172.55929565429688,
      "loss": 0.7032,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.10844483971595764,
      "rewards/margins": 0.0067189522087574005,
      "rewards/rejected": -0.11516381055116653,
      "step": 1200
    },
    {
      "epoch": 0.5174129353233831,
      "grad_norm": 1.5468394756317139,
      "learning_rate": 2.3624996759476285e-07,
      "logits/chosen": -0.9644728899002075,
      "logits/rejected": -0.9913946390151978,
      "logps/chosen": -178.84991455078125,
      "logps/rejected": -175.5871124267578,
      "loss": 0.6992,
      "rewards/accuracies": 0.5198437571525574,
      "rewards/chosen": -0.10432270169258118,
      "rewards/margins": 0.016916964203119278,
      "rewards/rejected": -0.12123966217041016,
      "step": 1300
    },
    {
      "epoch": 0.5572139303482587,
      "grad_norm": 1.2536767721176147,
      "learning_rate": 2.0522015093886614e-07,
      "logits/chosen": -0.9504435658454895,
      "logits/rejected": -0.9719184637069702,
      "logps/chosen": -178.3258056640625,
      "logps/rejected": -176.43231201171875,
      "loss": 0.6973,
      "rewards/accuracies": 0.514843761920929,
      "rewards/chosen": -0.09463655203580856,
      "rewards/margins": 0.02013176493346691,
      "rewards/rejected": -0.11476832628250122,
      "step": 1400
    },
    {
      "epoch": 0.5970149253731343,
      "grad_norm": 1.377833366394043,
      "learning_rate": 1.7488981696314154e-07,
      "logits/chosen": -0.9491727948188782,
      "logits/rejected": -0.97397780418396,
      "logps/chosen": -177.8011932373047,
      "logps/rejected": -174.43284606933594,
      "loss": 0.6986,
      "rewards/accuracies": 0.5153124928474426,
      "rewards/chosen": -0.08802775293588638,
      "rewards/margins": 0.014450294896960258,
      "rewards/rejected": -0.1024780347943306,
      "step": 1500
    },
    {
      "epoch": 0.6368159203980099,
      "grad_norm": 1.4358242750167847,
      "learning_rate": 1.4573274000458839e-07,
      "logits/chosen": -0.9687415957450867,
      "logits/rejected": -0.9858207106590271,
      "logps/chosen": -177.40037536621094,
      "logps/rejected": -175.6855010986328,
      "loss": 0.6988,
      "rewards/accuracies": 0.5137500166893005,
      "rewards/chosen": -0.08970288932323456,
      "rewards/margins": 0.016197971999645233,
      "rewards/rejected": -0.10590087622404099,
      "step": 1600
    },
    {
      "epoch": 0.6766169154228856,
      "grad_norm": 1.7245056629180908,
      "learning_rate": 1.1820436756391414e-07,
      "logits/chosen": -0.9455291032791138,
      "logits/rejected": -0.9768875241279602,
      "logps/chosen": -180.8955841064453,
      "logps/rejected": -176.17271423339844,
      "loss": 0.6959,
      "rewards/accuracies": 0.5231249928474426,
      "rewards/chosen": -0.08720911294221878,
      "rewards/margins": 0.021857038140296936,
      "rewards/rejected": -0.10906614363193512,
      "step": 1700
    },
    {
      "epoch": 0.7164179104477612,
      "grad_norm": 1.4436590671539307,
      "learning_rate": 9.273470599753375e-08,
      "logits/chosen": -0.9535448551177979,
      "logits/rejected": -0.968150794506073,
      "logps/chosen": -176.52774047851562,
      "logps/rejected": -175.44468688964844,
      "loss": 0.6951,
      "rewards/accuracies": 0.5285937786102295,
      "rewards/chosen": -0.08284671604633331,
      "rewards/margins": 0.022235672920942307,
      "rewards/rejected": -0.10508238524198532,
      "step": 1800
    },
    {
      "epoch": 0.7562189054726368,
      "grad_norm": 1.4405585527420044,
      "learning_rate": 6.972160361242119e-08,
      "logits/chosen": -0.9640732407569885,
      "logits/rejected": -0.9774580597877502,
      "logps/chosen": -174.50440979003906,
      "logps/rejected": -174.0882568359375,
      "loss": 0.6982,
      "rewards/accuracies": 0.5145312547683716,
      "rewards/chosen": -0.0837017223238945,
      "rewards/margins": 0.015292072668671608,
      "rewards/rejected": -0.09899380803108215,
      "step": 1900
    }
  ],
  "logging_steps": 100,
  "max_steps": 2512,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}