{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9969104016477858,
  "eval_steps": 100,
  "global_step": 121,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008238928939237899,
      "grad_norm": 3.6043033117037053,
      "learning_rate": 2.3076923076923076e-08,
      "logits/chosen": -0.971167802810669,
      "logits/rejected": -1.2423464059829712,
      "logps/chosen": -290.201171875,
      "logps/rejected": -263.9928894042969,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.082389289392379,
      "grad_norm": 3.4484942444647335,
      "learning_rate": 2.3076923076923078e-07,
      "logits/chosen": -0.9607771635055542,
      "logits/rejected": -0.9658682942390442,
      "logps/chosen": -330.0989685058594,
      "logps/rejected": -330.9446716308594,
      "loss": 0.6931,
      "rewards/accuracies": 0.4236111044883728,
      "rewards/chosen": 0.0003925244091078639,
      "rewards/margins": 4.308321513235569e-05,
      "rewards/rejected": 0.0003494411939755082,
      "step": 10
    },
    {
      "epoch": 0.164778578784758,
      "grad_norm": 3.5480284480156086,
      "learning_rate": 2.969010932648327e-07,
      "logits/chosen": -1.0245492458343506,
      "logits/rejected": -1.0337975025177002,
      "logps/chosen": -347.72344970703125,
      "logps/rejected": -345.3406066894531,
      "loss": 0.6907,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": 0.007954178377985954,
      "rewards/margins": 0.005248874891549349,
      "rewards/rejected": 0.00270530441775918,
      "step": 20
    },
    {
      "epoch": 0.24716786817713698,
      "grad_norm": 3.121263888821987,
      "learning_rate": 2.8203020867701666e-07,
      "logits/chosen": -0.8977586030960083,
      "logits/rejected": -1.0179085731506348,
      "logps/chosen": -304.6155700683594,
      "logps/rejected": -275.14447021484375,
      "loss": 0.6823,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": 0.025795791298151016,
      "rewards/margins": 0.025993749499320984,
      "rewards/rejected": -0.00019796025299001485,
      "step": 30
    },
    {
      "epoch": 0.329557157569516,
      "grad_norm": 2.6779190942219393,
      "learning_rate": 2.560660171779821e-07,
      "logits/chosen": -0.9125121235847473,
      "logits/rejected": -1.0349690914154053,
      "logps/chosen": -340.72442626953125,
      "logps/rejected": -313.9574890136719,
      "loss": 0.675,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": 0.02803516387939453,
      "rewards/margins": 0.04235466197133064,
      "rewards/rejected": -0.01431949995458126,
      "step": 40
    },
    {
      "epoch": 0.411946446961895,
      "grad_norm": 2.814818260153817,
      "learning_rate": 2.2119005546214604e-07,
      "logits/chosen": -0.9567445516586304,
      "logits/rejected": -0.9532279968261719,
      "logps/chosen": -272.302978515625,
      "logps/rejected": -295.3742370605469,
      "loss": 0.6649,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": 0.006898154504597187,
      "rewards/margins": 0.06003887206315994,
      "rewards/rejected": -0.05314071848988533,
      "step": 50
    },
    {
      "epoch": 0.49433573635427397,
      "grad_norm": 2.653170547007469,
      "learning_rate": 1.803326358498057e-07,
      "logits/chosen": -0.8227835893630981,
      "logits/rejected": -0.8951934576034546,
      "logps/chosen": -364.8822021484375,
      "logps/rejected": -332.07489013671875,
      "loss": 0.6563,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": 0.002518509980291128,
      "rewards/margins": 0.14299212396144867,
      "rewards/rejected": -0.1404736191034317,
      "step": 60
    },
    {
      "epoch": 0.576725025746653,
      "grad_norm": 2.7392936630658764,
      "learning_rate": 1.3692663858785124e-07,
      "logits/chosen": -0.8262346386909485,
      "logits/rejected": -0.9113919138908386,
      "logps/chosen": -337.5333251953125,
      "logps/rejected": -330.14459228515625,
      "loss": 0.6558,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.06830023974180222,
      "rewards/margins": 0.05756063386797905,
      "rewards/rejected": -0.12586086988449097,
      "step": 70
    },
    {
      "epoch": 0.659114315139032,
      "grad_norm": 4.467409541426095,
      "learning_rate": 9.461907790309735e-08,
      "logits/chosen": -0.8138895034790039,
      "logits/rejected": -0.9478031992912292,
      "logps/chosen": -370.1873779296875,
      "logps/rejected": -331.826904296875,
      "loss": 0.645,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.046670813113451004,
      "rewards/margins": 0.14053179323673248,
      "rewards/rejected": -0.187202587723732,
      "step": 80
    },
    {
      "epoch": 0.741503604531411,
      "grad_norm": 2.8035199165767315,
      "learning_rate": 5.6964676309761004e-08,
      "logits/chosen": -0.8749955296516418,
      "logits/rejected": -0.9215911030769348,
      "logps/chosen": -359.80511474609375,
      "logps/rejected": -361.1700134277344,
      "loss": 0.647,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.08991885185241699,
      "rewards/margins": 0.14792749285697937,
      "rewards/rejected": -0.23784634470939636,
      "step": 90
    },
    {
      "epoch": 0.82389289392379,
      "grad_norm": 3.018969560415363,
      "learning_rate": 2.7127193356651214e-08,
      "logits/chosen": -0.838853657245636,
      "logits/rejected": -0.981770396232605,
      "logps/chosen": -380.34307861328125,
      "logps/rejected": -369.41180419921875,
      "loss": 0.6411,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.10747754573822021,
      "rewards/margins": 0.17505204677581787,
      "rewards/rejected": -0.28252965211868286,
      "step": 100
    },
    {
      "epoch": 0.82389289392379,
      "eval_logits/chosen": -1.0462558269500732,
      "eval_logits/rejected": -1.1581745147705078,
      "eval_logps/chosen": -335.3811340332031,
      "eval_logps/rejected": -418.57818603515625,
      "eval_loss": 0.6494396924972534,
      "eval_rewards/accuracies": 0.5625,
      "eval_rewards/chosen": -0.17523552477359772,
      "eval_rewards/margins": 0.04425430670380592,
      "eval_rewards/rejected": -0.21948984265327454,
      "eval_runtime": 9.1337,
      "eval_samples_per_second": 109.485,
      "eval_steps_per_second": 0.876,
      "step": 100
    },
    {
      "epoch": 0.9062821833161689,
      "grad_norm": 2.971778165880155,
      "learning_rate": 7.613603464044916e-09,
      "logits/chosen": -0.8196293115615845,
      "logits/rejected": -0.9439749717712402,
      "logps/chosen": -354.5921630859375,
      "logps/rejected": -318.67962646484375,
      "loss": 0.6449,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.1414671391248703,
      "rewards/margins": 0.08098333328962326,
      "rewards/rejected": -0.22245049476623535,
      "step": 110
    },
    {
      "epoch": 0.9886714727085479,
      "grad_norm": 2.832084621380708,
      "learning_rate": 6.34574876699101e-11,
      "logits/chosen": -0.9086641073226929,
      "logits/rejected": -0.9848729372024536,
      "logps/chosen": -327.4648132324219,
      "logps/rejected": -343.4488525390625,
      "loss": 0.6437,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.10709067434072495,
      "rewards/margins": 0.1814233958721161,
      "rewards/rejected": -0.28851407766342163,
      "step": 120
    },
    {
      "epoch": 0.9969104016477858,
      "step": 121,
      "total_flos": 0.0,
      "train_loss": 0.6619468080110786,
      "train_runtime": 1288.9507,
      "train_samples_per_second": 48.206,
      "train_steps_per_second": 0.094
    }
  ],
  "logging_steps": 10,
  "max_steps": 121,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}