{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.971563981042654,
  "eval_steps": 100,
  "global_step": 104,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.018957345971563982,
      "grad_norm": 99.03995946284127,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": 163.72256469726562,
      "logits/rejected": 157.14466857910156,
      "logps/chosen": -692.1337890625,
      "logps/rejected": -783.4187622070312,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1895734597156398,
      "grad_norm": 111.11388003021844,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": 171.66250610351562,
      "logits/rejected": 172.8583221435547,
      "logps/chosen": -742.216064453125,
      "logps/rejected": -781.2522583007812,
      "loss": 0.7117,
      "rewards/accuracies": 0.4444444477558136,
      "rewards/chosen": 0.023412303999066353,
      "rewards/margins": 0.02978678233921528,
      "rewards/rejected": -0.006374475546181202,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 82.52463810141795,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": 163.12501525878906,
      "logits/rejected": 165.84164428710938,
      "logps/chosen": -709.082275390625,
      "logps/rejected": -749.2286376953125,
      "loss": 0.6534,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": 0.553031325340271,
      "rewards/margins": 0.20293028652668,
      "rewards/rejected": 0.35010096430778503,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 69.67853625499806,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": 184.1649932861328,
      "logits/rejected": 176.99354553222656,
      "logps/chosen": -710.6904907226562,
      "logps/rejected": -736.7250366210938,
      "loss": 0.5795,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": 1.0020567178726196,
      "rewards/margins": 0.513039767742157,
      "rewards/rejected": 0.4890168309211731,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 72.75906612350751,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": 167.30398559570312,
      "logits/rejected": 155.48980712890625,
      "logps/chosen": -697.3994140625,
      "logps/rejected": -700.686767578125,
      "loss": 0.5672,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": 0.41778701543807983,
      "rewards/margins": 0.858893871307373,
      "rewards/rejected": -0.44110679626464844,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 91.77276628667218,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": 182.02633666992188,
      "logits/rejected": 182.9945068359375,
      "logps/chosen": -771.2996826171875,
      "logps/rejected": -821.66943359375,
      "loss": 0.4967,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": 0.10474413633346558,
      "rewards/margins": 0.8699092864990234,
      "rewards/rejected": -0.7651651501655579,
      "step": 50
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 29.047916941624063,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": 174.08924865722656,
      "logits/rejected": 179.21060180664062,
      "logps/chosen": -725.1399536132812,
      "logps/rejected": -794.7996215820312,
      "loss": 0.2763,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": 0.714134693145752,
      "rewards/margins": 2.342179298400879,
      "rewards/rejected": -1.6280447244644165,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 28.169902323181958,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": 169.7666778564453,
      "logits/rejected": 167.54342651367188,
      "logps/chosen": -683.6437377929688,
      "logps/rejected": -783.1070556640625,
      "loss": 0.1734,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 0.8705979585647583,
      "rewards/margins": 2.8242533206939697,
      "rewards/rejected": -1.9536556005477905,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 24.1551722040215,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": 160.0974578857422,
      "logits/rejected": 172.2356414794922,
      "logps/chosen": -684.6734619140625,
      "logps/rejected": -806.4854125976562,
      "loss": 0.1605,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": 0.5792978405952454,
      "rewards/margins": 3.1283910274505615,
      "rewards/rejected": -2.549093246459961,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 25.406031573666652,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": 158.98464965820312,
      "logits/rejected": 170.3443145751953,
      "logps/chosen": -717.3196411132812,
      "logps/rejected": -813.6456298828125,
      "loss": 0.1493,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": 0.6181727647781372,
      "rewards/margins": 3.2944443225860596,
      "rewards/rejected": -2.6762712001800537,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 22.259191659621163,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": 163.0000762939453,
      "logits/rejected": 165.51370239257812,
      "logps/chosen": -720.561767578125,
      "logps/rejected": -794.7210693359375,
      "loss": 0.1424,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": 0.7237287759780884,
      "rewards/margins": 2.908165454864502,
      "rewards/rejected": -2.184436559677124,
      "step": 100
    },
    {
      "epoch": 1.8957345971563981,
      "eval_logits/chosen": 153.1355743408203,
      "eval_logits/rejected": 152.9660186767578,
      "eval_logps/chosen": -697.602294921875,
      "eval_logps/rejected": -720.2744750976562,
      "eval_loss": 0.47219032049179077,
      "eval_rewards/accuracies": 0.7395833134651184,
      "eval_rewards/chosen": -0.06579157710075378,
      "eval_rewards/margins": 1.2015198469161987,
      "eval_rewards/rejected": -1.267311453819275,
      "eval_runtime": 116.4984,
      "eval_samples_per_second": 6.438,
      "eval_steps_per_second": 0.206,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.3875045489806395,
      "train_runtime": 2331.3598,
      "train_samples_per_second": 5.791,
      "train_steps_per_second": 0.045
    }
  ],
  "logging_steps": 10,
  "max_steps": 104,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}