{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.996996996996997,
  "eval_steps": 66,
  "global_step": 166,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06006006006006006,
      "grad_norm": 34.52524185180664,
      "learning_rate": 9.936305732484077e-05,
      "logits/chosen": -0.5943359136581421,
      "logits/rejected": -0.6087890863418579,
      "logps/chosen": -16.725000381469727,
      "logps/rejected": -27.884374618530273,
      "loss": 0.6565,
      "nll_loss": 0.17890624701976776,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": -0.008593750186264515,
      "rewards/margins": 0.0693359375,
      "rewards/rejected": -0.07792969048023224,
      "step": 10
    },
    {
      "epoch": 0.12012012012012012,
      "grad_norm": 39.96049499511719,
      "learning_rate": 9.299363057324841e-05,
      "logits/chosen": -0.6298828125,
      "logits/rejected": -0.665820300579071,
      "logps/chosen": -21.274999618530273,
      "logps/rejected": -34.41875076293945,
      "loss": 0.5964,
      "nll_loss": 0.22294922173023224,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.4951171875,
      "rewards/margins": 0.8861328363418579,
      "rewards/rejected": -1.3815429210662842,
      "step": 20
    },
    {
      "epoch": 0.18018018018018017,
      "grad_norm": 26.530200958251953,
      "learning_rate": 8.662420382165606e-05,
      "logits/chosen": -0.60693359375,
      "logits/rejected": -0.6392577886581421,
      "logps/chosen": -20.003124237060547,
      "logps/rejected": -31.056249618530273,
      "loss": 0.6189,
      "nll_loss": 0.20683594048023224,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.4129394590854645,
      "rewards/margins": 0.711865246295929,
      "rewards/rejected": -1.1256835460662842,
      "step": 30
    },
    {
      "epoch": 0.24024024024024024,
      "grad_norm": 31.651872634887695,
      "learning_rate": 8.02547770700637e-05,
      "logits/chosen": -0.57861328125,
      "logits/rejected": -0.6119140386581421,
      "logps/chosen": -19.546875,
      "logps/rejected": -29.931249618530273,
      "loss": 0.6779,
      "nll_loss": 0.21328124403953552,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.4297851622104645,
      "rewards/margins": 0.8600097894668579,
      "rewards/rejected": -1.2890136241912842,
      "step": 40
    },
    {
      "epoch": 0.3003003003003003,
      "grad_norm": 26.210861206054688,
      "learning_rate": 7.388535031847134e-05,
      "logits/chosen": -0.5921875238418579,
      "logits/rejected": -0.62109375,
      "logps/chosen": -20.362499237060547,
      "logps/rejected": -33.70624923706055,
      "loss": 0.7017,
      "nll_loss": 0.208984375,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.5234375,
      "rewards/margins": 1.094140648841858,
      "rewards/rejected": -1.617285132408142,
      "step": 50
    },
    {
      "epoch": 0.36036036036036034,
      "grad_norm": 25.984054565429688,
      "learning_rate": 6.751592356687899e-05,
      "logits/chosen": -0.607226550579071,
      "logits/rejected": -0.625195324420929,
      "logps/chosen": -19.943750381469727,
      "logps/rejected": -33.96562576293945,
      "loss": 0.5971,
      "nll_loss": 0.2099609375,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.3338378965854645,
      "rewards/margins": 0.973828136920929,
      "rewards/rejected": -1.3087890148162842,
      "step": 60
    },
    {
      "epoch": 0.3963963963963964,
      "eval_logits/chosen": -0.6258928775787354,
      "eval_logits/rejected": -0.6188616156578064,
      "eval_logps/chosen": -21.77321434020996,
      "eval_logps/rejected": -34.83571243286133,
      "eval_loss": 0.6648827195167542,
      "eval_nll_loss": 0.20853795111179352,
      "eval_rewards/accuracies": 0.6821428537368774,
      "eval_rewards/chosen": -0.4113839268684387,
      "eval_rewards/margins": 1.0587611198425293,
      "eval_rewards/rejected": -1.470479965209961,
      "eval_runtime": 26.32,
      "eval_samples_per_second": 10.562,
      "eval_steps_per_second": 1.33,
      "step": 66
    },
    {
      "epoch": 0.42042042042042044,
      "grad_norm": 26.273643493652344,
      "learning_rate": 6.114649681528662e-05,
      "logits/chosen": -0.618945300579071,
      "logits/rejected": -0.6214843988418579,
      "logps/chosen": -18.078125,
      "logps/rejected": -30.125,
      "loss": 0.5691,
      "nll_loss": 0.19150391221046448,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.13154296576976776,
      "rewards/margins": 1.20947265625,
      "rewards/rejected": -1.339257836341858,
      "step": 70
    },
    {
      "epoch": 0.4804804804804805,
      "grad_norm": 21.918045043945312,
      "learning_rate": 5.477707006369427e-05,
      "logits/chosen": -0.603515625,
      "logits/rejected": -0.615527331829071,
      "logps/chosen": -20.246875762939453,
      "logps/rejected": -33.287498474121094,
      "loss": 0.5856,
      "nll_loss": 0.21635742485523224,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.5206054449081421,
      "rewards/margins": 1.24072265625,
      "rewards/rejected": -1.760839819908142,
      "step": 80
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 32.28929901123047,
      "learning_rate": 4.840764331210191e-05,
      "logits/chosen": -0.5863281488418579,
      "logits/rejected": -0.635546863079071,
      "logps/chosen": -20.846874237060547,
      "logps/rejected": -33.584373474121094,
      "loss": 0.6504,
      "nll_loss": 0.2353515625,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.680957019329071,
      "rewards/margins": 1.7887694835662842,
      "rewards/rejected": -2.4681639671325684,
      "step": 90
    },
    {
      "epoch": 0.6006006006006006,
      "grad_norm": 42.50503921508789,
      "learning_rate": 4.2038216560509556e-05,
      "logits/chosen": -0.615429699420929,
      "logits/rejected": -0.605273425579071,
      "logps/chosen": -20.068750381469727,
      "logps/rejected": -31.799999237060547,
      "loss": 0.6865,
      "nll_loss": 0.206787109375,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.53125,
      "rewards/margins": 1.4816405773162842,
      "rewards/rejected": -2.0130858421325684,
      "step": 100
    },
    {
      "epoch": 0.6606606606606606,
      "grad_norm": 35.92078399658203,
      "learning_rate": 3.56687898089172e-05,
      "logits/chosen": -0.614941418170929,
      "logits/rejected": -0.617382824420929,
      "logps/chosen": -20.15625,
      "logps/rejected": -36.556251525878906,
      "loss": 0.6166,
      "nll_loss": 0.20083007216453552,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.663525402545929,
      "rewards/margins": 1.5426757335662842,
      "rewards/rejected": -2.207714796066284,
      "step": 110
    },
    {
      "epoch": 0.7207207207207207,
      "grad_norm": 35.85774612426758,
      "learning_rate": 2.929936305732484e-05,
      "logits/chosen": -0.6103515625,
      "logits/rejected": -0.6181640625,
      "logps/chosen": -21.934375762939453,
      "logps/rejected": -35.60625076293945,
      "loss": 0.5358,
      "nll_loss": 0.21962890028953552,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.369140625,
      "rewards/margins": 1.749414086341858,
      "rewards/rejected": -2.1181640625,
      "step": 120
    },
    {
      "epoch": 0.7807807807807807,
      "grad_norm": 29.95037078857422,
      "learning_rate": 2.2929936305732484e-05,
      "logits/chosen": -0.596875011920929,
      "logits/rejected": -0.6224609613418579,
      "logps/chosen": -18.59375,
      "logps/rejected": -35.493751525878906,
      "loss": 0.4837,
      "nll_loss": 0.19125977158546448,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.36542969942092896,
      "rewards/margins": 2.2035155296325684,
      "rewards/rejected": -2.5687499046325684,
      "step": 130
    },
    {
      "epoch": 0.7927927927927928,
      "eval_logits/chosen": -0.6104352474212646,
      "eval_logits/rejected": -0.6017857193946838,
      "eval_logps/chosen": -22.016071319580078,
      "eval_logps/rejected": -35.73214340209961,
      "eval_loss": 0.6540421843528748,
      "eval_nll_loss": 0.21071428060531616,
      "eval_rewards/accuracies": 0.6714285612106323,
      "eval_rewards/chosen": -0.5337053537368774,
      "eval_rewards/margins": 1.3877789974212646,
      "eval_rewards/rejected": -1.9204241037368774,
      "eval_runtime": 26.2394,
      "eval_samples_per_second": 10.595,
      "eval_steps_per_second": 1.334,
      "step": 132
    },
    {
      "epoch": 0.8408408408408409,
      "grad_norm": 17.50779914855957,
      "learning_rate": 1.6560509554140128e-05,
      "logits/chosen": -0.621874988079071,
      "logits/rejected": -0.6468750238418579,
      "logps/chosen": -18.740625381469727,
      "logps/rejected": -36.33124923706055,
      "loss": 0.5761,
      "nll_loss": 0.20664063096046448,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.4644531309604645,
      "rewards/margins": 1.7863280773162842,
      "rewards/rejected": -2.2529296875,
      "step": 140
    },
    {
      "epoch": 0.9009009009009009,
      "grad_norm": 24.695812225341797,
      "learning_rate": 1.0191082802547772e-05,
      "logits/chosen": -0.613964855670929,
      "logits/rejected": -0.6097656488418579,
      "logps/chosen": -17.393749237060547,
      "logps/rejected": -33.525001525878906,
      "loss": 0.4594,
      "nll_loss": 0.19106444716453552,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -0.20957031846046448,
      "rewards/margins": 2.38232421875,
      "rewards/rejected": -2.591796875,
      "step": 150
    },
    {
      "epoch": 0.960960960960961,
      "grad_norm": 17.522380828857422,
      "learning_rate": 3.821656050955414e-06,
      "logits/chosen": -0.612500011920929,
      "logits/rejected": -0.6226562261581421,
      "logps/chosen": -19.4375,
      "logps/rejected": -35.01250076293945,
      "loss": 0.5153,
      "nll_loss": 0.20332030951976776,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.4234375059604645,
      "rewards/margins": 2.30859375,
      "rewards/rejected": -2.732714891433716,
      "step": 160
    },
    {
      "epoch": 0.996996996996997,
      "step": 166,
      "total_flos": 0.0,
      "train_loss": 0.6033119293580572,
      "train_runtime": 697.6675,
      "train_samples_per_second": 3.817,
      "train_steps_per_second": 0.238
    }
  ],
  "logging_steps": 10,
  "max_steps": 166,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 66,
  "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": 1,
  "trial_name": null,
  "trial_params": null
}