{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9965156794425087,
  "eval_steps": 50,
  "global_step": 143,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06968641114982578,
      "grad_norm": 8.18808377346908,
      "learning_rate": 3.333333333333333e-07,
      "logits/chosen": -2.7029919624328613,
      "logits/rejected": -2.729884386062622,
      "logps/chosen": -301.84100341796875,
      "logps/rejected": -331.34771728515625,
      "loss": 0.6931,
      "rewards/accuracies": 0.4124999940395355,
      "rewards/chosen": 0.00025743356673046947,
      "rewards/margins": 0.00021156822913326323,
      "rewards/rejected": 4.586534487316385e-05,
      "step": 10
    },
    {
      "epoch": 0.13937282229965156,
      "grad_norm": 7.811499447648029,
      "learning_rate": 4.981198836496775e-07,
      "logits/chosen": -2.7489161491394043,
      "logits/rejected": -2.737814426422119,
      "logps/chosen": -320.05413818359375,
      "logps/rejected": -313.6223449707031,
      "loss": 0.6924,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.004671993665397167,
      "rewards/margins": 0.0004203595162834972,
      "rewards/rejected": 0.004251633305102587,
      "step": 20
    },
    {
      "epoch": 0.20905923344947736,
      "grad_norm": 9.855807812241265,
      "learning_rate": 4.832481997086846e-07,
      "logits/chosen": -2.6600353717803955,
      "logits/rejected": -2.6442506313323975,
      "logps/chosen": -291.5880126953125,
      "logps/rejected": -304.8880615234375,
      "loss": 0.6907,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": 0.046477120369672775,
      "rewards/margins": -0.0016271257773041725,
      "rewards/rejected": 0.04810425639152527,
      "step": 30
    },
    {
      "epoch": 0.2787456445993031,
      "grad_norm": 7.883815856532723,
      "learning_rate": 4.543962032878959e-07,
      "logits/chosen": -2.625170946121216,
      "logits/rejected": -2.6397366523742676,
      "logps/chosen": -286.61895751953125,
      "logps/rejected": -294.7152404785156,
      "loss": 0.6849,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": 0.036045268177986145,
      "rewards/margins": 0.013393419794738293,
      "rewards/rejected": 0.02265184372663498,
      "step": 40
    },
    {
      "epoch": 0.34843205574912894,
      "grad_norm": 8.729035466800674,
      "learning_rate": 4.1329321073844413e-07,
      "logits/chosen": -2.594357967376709,
      "logits/rejected": -2.6037957668304443,
      "logps/chosen": -331.63134765625,
      "logps/rejected": -313.75775146484375,
      "loss": 0.6816,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.029668737202882767,
      "rewards/margins": 0.02038290724158287,
      "rewards/rejected": -0.05005164071917534,
      "step": 50
    },
    {
      "epoch": 0.34843205574912894,
      "eval_logits/chosen": -2.5709125995635986,
      "eval_logits/rejected": -2.5365915298461914,
      "eval_logps/chosen": -263.3412780761719,
      "eval_logps/rejected": -268.15484619140625,
      "eval_loss": 0.672498881816864,
      "eval_rewards/accuracies": 0.58984375,
      "eval_rewards/chosen": -0.0071137575432658195,
      "eval_rewards/margins": 0.04781381040811539,
      "eval_rewards/rejected": -0.05492757260799408,
      "eval_runtime": 101.2981,
      "eval_samples_per_second": 19.744,
      "eval_steps_per_second": 0.316,
      "step": 50
    },
    {
      "epoch": 0.4181184668989547,
      "grad_norm": 9.255478757143024,
      "learning_rate": 3.624028324136517e-07,
      "logits/chosen": -2.6415677070617676,
      "logits/rejected": -2.614701509475708,
      "logps/chosen": -305.6767883300781,
      "logps/rejected": -307.7297668457031,
      "loss": 0.679,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.0592481791973114,
      "rewards/margins": 0.0437784343957901,
      "rewards/rejected": -0.1030266135931015,
      "step": 60
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 9.856902330832456,
      "learning_rate": 3.047753100392174e-07,
      "logits/chosen": -2.598100185394287,
      "logits/rejected": -2.59047532081604,
      "logps/chosen": -330.283203125,
      "logps/rejected": -311.8639831542969,
      "loss": 0.6799,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.04586242884397507,
      "rewards/margins": 0.05955269932746887,
      "rewards/rejected": -0.10541512817144394,
      "step": 70
    },
    {
      "epoch": 0.5574912891986062,
      "grad_norm": 10.552750099759244,
      "learning_rate": 2.4386469286927194e-07,
      "logits/chosen": -2.611276149749756,
      "logits/rejected": -2.5819907188415527,
      "logps/chosen": -285.20318603515625,
      "logps/rejected": -287.8341369628906,
      "loss": 0.6733,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": 0.030143652111291885,
      "rewards/margins": 0.07110844552516937,
      "rewards/rejected": -0.04096478223800659,
      "step": 80
    },
    {
      "epoch": 0.627177700348432,
      "grad_norm": 14.503008300206526,
      "learning_rate": 1.8332181063127542e-07,
      "logits/chosen": -2.5344879627227783,
      "logits/rejected": -2.5114903450012207,
      "logps/chosen": -313.04052734375,
      "logps/rejected": -312.25811767578125,
      "loss": 0.6749,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.07387819141149521,
      "rewards/margins": 0.06775785237550735,
      "rewards/rejected": -0.14163604378700256,
      "step": 90
    },
    {
      "epoch": 0.6968641114982579,
      "grad_norm": 11.233854788459887,
      "learning_rate": 1.26775451942554e-07,
      "logits/chosen": -2.5010740756988525,
      "logits/rejected": -2.4948906898498535,
      "logps/chosen": -304.60186767578125,
      "logps/rejected": -325.60968017578125,
      "loss": 0.6723,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.10407546907663345,
      "rewards/margins": 0.027523279190063477,
      "rewards/rejected": -0.13159874081611633,
      "step": 100
    },
    {
      "epoch": 0.6968641114982579,
      "eval_logits/chosen": -2.495295286178589,
      "eval_logits/rejected": -2.459512948989868,
      "eval_logps/chosen": -271.3895568847656,
      "eval_logps/rejected": -278.11297607421875,
      "eval_loss": 0.6687415242195129,
      "eval_rewards/accuracies": 0.57421875,
      "eval_rewards/chosen": -0.08759630471467972,
      "eval_rewards/margins": 0.0669122040271759,
      "eval_rewards/rejected": -0.15450850129127502,
      "eval_runtime": 102.2188,
      "eval_samples_per_second": 19.566,
      "eval_steps_per_second": 0.313,
      "step": 100
    },
    {
      "epoch": 0.7665505226480837,
      "grad_norm": 11.106233348415344,
      "learning_rate": 7.761486381573326e-08,
      "logits/chosen": -2.547499179840088,
      "logits/rejected": -2.5436906814575195,
      "logps/chosen": -311.0939025878906,
      "logps/rejected": -317.546875,
      "loss": 0.6789,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.07100532948970795,
      "rewards/margins": 0.048421461135149,
      "rewards/rejected": -0.11942678689956665,
      "step": 110
    },
    {
      "epoch": 0.8362369337979094,
      "grad_norm": 8.671494044278699,
      "learning_rate": 3.878660868757322e-08,
      "logits/chosen": -2.4711318016052246,
      "logits/rejected": -2.4641852378845215,
      "logps/chosen": -250.763916015625,
      "logps/rejected": -268.88031005859375,
      "loss": 0.6795,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.021220296621322632,
      "rewards/margins": 0.010579629801213741,
      "rewards/rejected": -0.03179992362856865,
      "step": 120
    },
    {
      "epoch": 0.9059233449477352,
      "grad_norm": 16.493475346448555,
      "learning_rate": 1.261795485174083e-08,
      "logits/chosen": -2.546675205230713,
      "logits/rejected": -2.548250675201416,
      "logps/chosen": -298.49188232421875,
      "logps/rejected": -296.81341552734375,
      "loss": 0.68,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": 0.026115726679563522,
      "rewards/margins": 0.03052101656794548,
      "rewards/rejected": -0.004405293148010969,
      "step": 130
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 10.645354039301385,
      "learning_rate": 6.773858303274482e-10,
      "logits/chosen": -2.5520262718200684,
      "logits/rejected": -2.539773941040039,
      "logps/chosen": -326.08477783203125,
      "logps/rejected": -305.88226318359375,
      "loss": 0.6766,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.00047526060370728374,
      "rewards/margins": 0.04427434504032135,
      "rewards/rejected": -0.044749610126018524,
      "step": 140
    },
    {
      "epoch": 0.9965156794425087,
      "step": 143,
      "total_flos": 0.0,
      "train_loss": 0.6818205426622937,
      "train_runtime": 3740.6507,
      "train_samples_per_second": 4.903,
      "train_steps_per_second": 0.038
    }
  ],
  "logging_steps": 10,
  "max_steps": 143,
  "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": 8,
  "trial_name": null,
  "trial_params": null
}