{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9968652037617555,
  "eval_steps": 500,
  "global_step": 159,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.125e-08,
      "logits/chosen": -2.6679039001464844,
      "logits/rejected": -2.5546512603759766,
      "logps/chosen": -164.941650390625,
      "logps/pi_response": -79.30451965332031,
      "logps/ref_response": -79.30451965332031,
      "logps/rejected": -140.9091339111328,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": -2.6874029636383057,
      "logits/rejected": -2.6898481845855713,
      "logps/chosen": -237.28390502929688,
      "logps/pi_response": -119.58097076416016,
      "logps/ref_response": -118.72599029541016,
      "logps/rejected": -235.39962768554688,
      "loss": 0.6916,
      "rewards/accuracies": 0.4166666567325592,
      "rewards/chosen": -0.011355193331837654,
      "rewards/margins": 0.00026273002731613815,
      "rewards/rejected": -0.011617923155426979,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.990353313429303e-07,
      "logits/chosen": -2.7030341625213623,
      "logits/rejected": -2.6754679679870605,
      "logps/chosen": -269.548095703125,
      "logps/pi_response": -143.17662048339844,
      "logps/ref_response": -130.33164978027344,
      "logps/rejected": -247.9852752685547,
      "loss": 0.6775,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.16177864372730255,
      "rewards/margins": 0.03440982848405838,
      "rewards/rejected": -0.19618846476078033,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.882681251368548e-07,
      "logits/chosen": -2.5051536560058594,
      "logits/rejected": -2.44065260887146,
      "logps/chosen": -242.4348907470703,
      "logps/pi_response": -128.1033935546875,
      "logps/ref_response": -112.08052825927734,
      "logps/rejected": -250.30142211914062,
      "loss": 0.6658,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.16098998486995697,
      "rewards/margins": 0.12281806766986847,
      "rewards/rejected": -0.28380805253982544,
      "step": 30
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.6604720940421207e-07,
      "logits/chosen": -2.3018674850463867,
      "logits/rejected": -2.2825875282287598,
      "logps/chosen": -250.5218963623047,
      "logps/pi_response": -155.37045288085938,
      "logps/ref_response": -110.90777587890625,
      "logps/rejected": -318.40362548828125,
      "loss": 0.6482,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.4296089708805084,
      "rewards/margins": 0.17231182754039764,
      "rewards/rejected": -0.6019207239151001,
      "step": 40
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3344075855595097e-07,
      "logits/chosen": -2.2006545066833496,
      "logits/rejected": -2.175128936767578,
      "logps/chosen": -277.812255859375,
      "logps/pi_response": -171.28018188476562,
      "logps/ref_response": -117.65016174316406,
      "logps/rejected": -289.81500244140625,
      "loss": 0.6416,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.4961293339729309,
      "rewards/margins": 0.10620589554309845,
      "rewards/rejected": -0.6023353338241577,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.920161866827889e-07,
      "logits/chosen": -2.1257805824279785,
      "logits/rejected": -2.091686487197876,
      "logps/chosen": -265.3650817871094,
      "logps/pi_response": -174.8240203857422,
      "logps/ref_response": -108.0089340209961,
      "logps/rejected": -296.00775146484375,
      "loss": 0.6372,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.4514217972755432,
      "rewards/margins": 0.30198341608047485,
      "rewards/rejected": -0.7534051537513733,
      "step": 60
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.4376480090239047e-07,
      "logits/chosen": -2.079555034637451,
      "logits/rejected": -2.030536651611328,
      "logps/chosen": -294.84588623046875,
      "logps/pi_response": -209.66567993164062,
      "logps/ref_response": -112.66800689697266,
      "logps/rejected": -317.8525085449219,
      "loss": 0.6129,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.8479757308959961,
      "rewards/margins": 0.3113170564174652,
      "rewards/rejected": -1.1592928171157837,
      "step": 70
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.910060778827554e-07,
      "logits/chosen": -2.0623815059661865,
      "logits/rejected": -2.004610061645508,
      "logps/chosen": -285.8151550292969,
      "logps/pi_response": -202.29714965820312,
      "logps/ref_response": -101.95613098144531,
      "logps/rejected": -329.02276611328125,
      "loss": 0.6258,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.8381752967834473,
      "rewards/margins": 0.27770641446113586,
      "rewards/rejected": -1.1158816814422607,
      "step": 80
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.3627616503391812e-07,
      "logits/chosen": -2.119082450866699,
      "logits/rejected": -2.078995704650879,
      "logps/chosen": -286.7127990722656,
      "logps/pi_response": -184.21145629882812,
      "logps/ref_response": -100.02299499511719,
      "logps/rejected": -306.8168029785156,
      "loss": 0.6152,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.6839859485626221,
      "rewards/margins": 0.3100510239601135,
      "rewards/rejected": -0.9940369725227356,
      "step": 90
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8220596619089573e-07,
      "logits/chosen": -2.054029941558838,
      "logits/rejected": -2.045511484146118,
      "logps/chosen": -293.54083251953125,
      "logps/pi_response": -204.90469360351562,
      "logps/ref_response": -112.04389953613281,
      "logps/rejected": -355.5297546386719,
      "loss": 0.5937,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.7882789373397827,
      "rewards/margins": 0.3712361752986908,
      "rewards/rejected": -1.1595150232315063,
      "step": 100
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3139467229135998e-07,
      "logits/chosen": -2.1169228553771973,
      "logits/rejected": -2.1269078254699707,
      "logps/chosen": -281.44061279296875,
      "logps/pi_response": -183.658935546875,
      "logps/ref_response": -100.44270324707031,
      "logps/rejected": -322.4181213378906,
      "loss": 0.6039,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.7017567753791809,
      "rewards/margins": 0.32792752981185913,
      "rewards/rejected": -1.02968430519104,
      "step": 110
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.628481651367875e-08,
      "logits/chosen": -2.1294326782226562,
      "logits/rejected": -2.056295871734619,
      "logps/chosen": -314.05352783203125,
      "logps/pi_response": -221.9539337158203,
      "logps/ref_response": -129.4988250732422,
      "logps/rejected": -340.3128662109375,
      "loss": 0.5942,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.7849725484848022,
      "rewards/margins": 0.2975076138973236,
      "rewards/rejected": -1.0824801921844482,
      "step": 120
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.904486005914027e-08,
      "logits/chosen": -2.159844398498535,
      "logits/rejected": -2.1421799659729004,
      "logps/chosen": -285.5159606933594,
      "logps/pi_response": -207.9140625,
      "logps/ref_response": -112.079345703125,
      "logps/rejected": -348.6537780761719,
      "loss": 0.5918,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.8335883021354675,
      "rewards/margins": 0.3036562204360962,
      "rewards/rejected": -1.1372445821762085,
      "step": 130
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1464952759020856e-08,
      "logits/chosen": -2.1429715156555176,
      "logits/rejected": -2.089409589767456,
      "logps/chosen": -312.14837646484375,
      "logps/pi_response": -233.33963012695312,
      "logps/ref_response": -119.2865219116211,
      "logps/rejected": -356.63690185546875,
      "loss": 0.5961,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.8318526148796082,
      "rewards/margins": 0.4721902310848236,
      "rewards/rejected": -1.3040426969528198,
      "step": 140
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.8708793644441086e-09,
      "logits/chosen": -2.0958361625671387,
      "logits/rejected": -2.0661659240722656,
      "logps/chosen": -289.91314697265625,
      "logps/pi_response": -198.4707489013672,
      "logps/ref_response": -101.13041687011719,
      "logps/rejected": -321.5401306152344,
      "loss": 0.5714,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.752092182636261,
      "rewards/margins": 0.3856619894504547,
      "rewards/rejected": -1.1377540826797485,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 159,
      "total_flos": 0.0,
      "train_loss": 0.6240486348949889,
      "train_runtime": 4176.0906,
      "train_samples_per_second": 4.88,
      "train_steps_per_second": 0.038
    }
  ],
  "logging_steps": 10,
  "max_steps": 159,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}