{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.99581589958159,
  "eval_steps": 500,
  "global_step": 119,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.166666666666666e-08,
      "logits/chosen": -2.3755712509155273,
      "logits/rejected": -2.380011558532715,
      "logps/chosen": -396.8262023925781,
      "logps/pi_response": -147.47637939453125,
      "logps/ref_response": -147.47637939453125,
      "logps/rejected": -480.5025634765625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -2.306025266647339,
      "logits/rejected": -2.296987533569336,
      "logps/chosen": -434.2967529296875,
      "logps/pi_response": -168.16778564453125,
      "logps/ref_response": -168.15762329101562,
      "logps/rejected": -477.62066650390625,
      "loss": 0.6894,
      "rewards/accuracies": 0.5208333134651184,
      "rewards/chosen": -0.03276034817099571,
      "rewards/margins": 0.006516492459923029,
      "rewards/rejected": -0.0392768420279026,
      "step": 10
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.931352528237397e-07,
      "logits/chosen": -2.1829891204833984,
      "logits/rejected": -2.1276068687438965,
      "logps/chosen": -456.19000244140625,
      "logps/pi_response": -159.04637145996094,
      "logps/ref_response": -153.50245666503906,
      "logps/rejected": -582.487548828125,
      "loss": 0.6432,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.5500961542129517,
      "rewards/margins": 0.2686466574668884,
      "rewards/rejected": -0.8187428712844849,
      "step": 20
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.658920803689553e-07,
      "logits/chosen": -2.125579833984375,
      "logits/rejected": -2.0711355209350586,
      "logps/chosen": -478.7774353027344,
      "logps/pi_response": -221.3973846435547,
      "logps/ref_response": -205.2119903564453,
      "logps/rejected": -575.614990234375,
      "loss": 0.6322,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.5594707131385803,
      "rewards/margins": 0.3740437924861908,
      "rewards/rejected": -0.9335145950317383,
      "step": 30
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.201712553872657e-07,
      "logits/chosen": -2.1675801277160645,
      "logits/rejected": -2.1256229877471924,
      "logps/chosen": -491.529052734375,
      "logps/pi_response": -201.5789794921875,
      "logps/ref_response": -166.27740478515625,
      "logps/rejected": -564.7361450195312,
      "loss": 0.6135,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.7727736830711365,
      "rewards/margins": 0.25983017683029175,
      "rewards/rejected": -1.0326038599014282,
      "step": 40
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.598859066780754e-07,
      "logits/chosen": -2.0738723278045654,
      "logits/rejected": -2.055086851119995,
      "logps/chosen": -498.4329528808594,
      "logps/pi_response": -215.8629913330078,
      "logps/ref_response": -181.7102508544922,
      "logps/rejected": -572.1842041015625,
      "loss": 0.5976,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.6373351812362671,
      "rewards/margins": 0.30541670322418213,
      "rewards/rejected": -0.9427520036697388,
      "step": 50
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9019570347986706e-07,
      "logits/chosen": -2.0982589721679688,
      "logits/rejected": -2.0577781200408936,
      "logps/chosen": -498.26055908203125,
      "logps/pi_response": -217.1796875,
      "logps/ref_response": -177.24761962890625,
      "logps/rejected": -615.952880859375,
      "loss": 0.6182,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.8855204582214355,
      "rewards/margins": 0.4653972089290619,
      "rewards/rejected": -1.3509178161621094,
      "step": 60
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1706525253979534e-07,
      "logits/chosen": -2.0901355743408203,
      "logits/rejected": -2.017996072769165,
      "logps/chosen": -492.5106506347656,
      "logps/pi_response": -234.0646514892578,
      "logps/ref_response": -189.48513793945312,
      "logps/rejected": -597.0025024414062,
      "loss": 0.5864,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.46928316354751587,
      "rewards/margins": 0.5200775265693665,
      "rewards/rejected": -0.9893606305122375,
      "step": 70
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4675360263490295e-07,
      "logits/chosen": -1.9807932376861572,
      "logits/rejected": -1.936804175376892,
      "logps/chosen": -479.35595703125,
      "logps/pi_response": -225.93392944335938,
      "logps/ref_response": -176.03411865234375,
      "logps/rejected": -591.875244140625,
      "loss": 0.5799,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.7031043171882629,
      "rewards/margins": 0.40889111161231995,
      "rewards/rejected": -1.1119954586029053,
      "step": 80
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.527854855097224e-08,
      "logits/chosen": -1.8924801349639893,
      "logits/rejected": -1.8593276739120483,
      "logps/chosen": -575.15185546875,
      "logps/pi_response": -248.3000946044922,
      "logps/ref_response": -182.91275024414062,
      "logps/rejected": -703.632080078125,
      "loss": 0.5781,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.2331480979919434,
      "rewards/margins": 0.5850399136543274,
      "rewards/rejected": -1.8181877136230469,
      "step": 90
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.790158337517127e-08,
      "logits/chosen": -1.9359309673309326,
      "logits/rejected": -1.9223928451538086,
      "logps/chosen": -529.174072265625,
      "logps/pi_response": -227.7006378173828,
      "logps/ref_response": -164.98428344726562,
      "logps/rejected": -648.9022216796875,
      "loss": 0.5683,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -1.045077919960022,
      "rewards/margins": 0.5439373850822449,
      "rewards/rejected": -1.589015245437622,
      "step": 100
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.677580722139671e-09,
      "logits/chosen": -1.8646703958511353,
      "logits/rejected": -1.8161742687225342,
      "logps/chosen": -512.7567138671875,
      "logps/pi_response": -230.37588500976562,
      "logps/ref_response": -167.1942138671875,
      "logps/rejected": -604.599609375,
      "loss": 0.5676,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.034223198890686,
      "rewards/margins": 0.4625032842159271,
      "rewards/rejected": -1.496726632118225,
      "step": 110
    },
    {
      "epoch": 1.0,
      "step": 119,
      "total_flos": 0.0,
      "train_loss": 0.6084108673223928,
      "train_runtime": 3108.4145,
      "train_samples_per_second": 4.917,
      "train_steps_per_second": 0.038
    }
  ],
  "logging_steps": 10,
  "max_steps": 119,
  "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
}