{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5333333333333333,
  "eval_steps": 50,
  "global_step": 600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.044444444444444446,
      "grad_norm": 49.98692321777344,
      "learning_rate": 5e-07,
      "logits/chosen": -1.6455594301223755,
      "logits/rejected": -1.8609813451766968,
      "logps/chosen": -69.99601745605469,
      "logps/rejected": -28.68402099609375,
      "loss": 0.4811,
      "rewards/accuracies": 0.9200000166893005,
      "rewards/chosen": 0.43002215027809143,
      "rewards/margins": 0.5248908400535583,
      "rewards/rejected": -0.09486865252256393,
      "step": 50
    },
    {
      "epoch": 0.044444444444444446,
      "eval_logits/chosen": -1.683065414428711,
      "eval_logits/rejected": -1.8381606340408325,
      "eval_logps/chosen": -65.2982406616211,
      "eval_logps/rejected": -31.1202335357666,
      "eval_loss": 0.26570644974708557,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 0.9695585370063782,
      "eval_rewards/margins": 1.199601650238037,
      "eval_rewards/rejected": -0.23004300892353058,
      "eval_runtime": 547.5928,
      "eval_samples_per_second": 0.51,
      "eval_steps_per_second": 0.51,
      "step": 50
    },
    {
      "epoch": 0.08888888888888889,
      "grad_norm": 18.30071449279785,
      "learning_rate": 5e-07,
      "logits/chosen": -1.7492402791976929,
      "logits/rejected": -1.8840913772583008,
      "logps/chosen": -62.02247619628906,
      "logps/rejected": -31.541664123535156,
      "loss": 0.1648,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.368556022644043,
      "rewards/margins": 1.8034323453903198,
      "rewards/rejected": -0.4348761737346649,
      "step": 100
    },
    {
      "epoch": 0.08888888888888889,
      "eval_logits/chosen": -1.717586636543274,
      "eval_logits/rejected": -1.840760588645935,
      "eval_logps/chosen": -57.23891067504883,
      "eval_logps/rejected": -35.770687103271484,
      "eval_loss": 0.0847974419593811,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.77549147605896,
      "eval_rewards/margins": 2.4705796241760254,
      "eval_rewards/rejected": -0.6950880885124207,
      "eval_runtime": 547.9305,
      "eval_samples_per_second": 0.509,
      "eval_steps_per_second": 0.509,
      "step": 100
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 6.001527309417725,
      "learning_rate": 5e-07,
      "logits/chosen": -1.753422498703003,
      "logits/rejected": -1.856553316116333,
      "logps/chosen": -55.98763656616211,
      "logps/rejected": -39.17717361450195,
      "loss": 0.0522,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.045351266860962,
      "rewards/margins": 3.051741361618042,
      "rewards/rejected": -1.00639009475708,
      "step": 150
    },
    {
      "epoch": 0.13333333333333333,
      "eval_logits/chosen": -1.7338056564331055,
      "eval_logits/rejected": -1.838354229927063,
      "eval_logps/chosen": -51.332515716552734,
      "eval_logps/rejected": -42.16679382324219,
      "eval_loss": 0.026723211631178856,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.366131067276001,
      "eval_rewards/margins": 3.7008297443389893,
      "eval_rewards/rejected": -1.3346984386444092,
      "eval_runtime": 548.2166,
      "eval_samples_per_second": 0.509,
      "eval_steps_per_second": 0.509,
      "step": 150
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 3.3345212936401367,
      "learning_rate": 5e-07,
      "logits/chosen": -1.7408527135849,
      "logits/rejected": -1.8468482494354248,
      "logps/chosen": -51.4506721496582,
      "logps/rejected": -46.23123550415039,
      "loss": 0.0157,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.529080867767334,
      "rewards/margins": 4.344725131988525,
      "rewards/rejected": -1.8156436681747437,
      "step": 200
    },
    {
      "epoch": 0.17777777777777778,
      "eval_logits/chosen": -1.744242787361145,
      "eval_logits/rejected": -1.8309202194213867,
      "eval_logps/chosen": -47.80553436279297,
      "eval_logps/rejected": -54.1657600402832,
      "eval_loss": 0.006161259487271309,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.7188286781311035,
      "eval_rewards/margins": 5.253425121307373,
      "eval_rewards/rejected": -2.5345959663391113,
      "eval_runtime": 548.15,
      "eval_samples_per_second": 0.509,
      "eval_steps_per_second": 0.509,
      "step": 200
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 0.14823301136493683,
      "learning_rate": 5e-07,
      "logits/chosen": -1.747748851776123,
      "logits/rejected": -1.8711057901382446,
      "logps/chosen": -48.346378326416016,
      "logps/rejected": -64.187744140625,
      "loss": 0.0026,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.810575485229492,
      "rewards/margins": 6.410886764526367,
      "rewards/rejected": -3.600311040878296,
      "step": 250
    },
    {
      "epoch": 0.2222222222222222,
      "eval_logits/chosen": -1.7560131549835205,
      "eval_logits/rejected": -1.8139592409133911,
      "eval_logps/chosen": -46.13553237915039,
      "eval_logps/rejected": -74.509765625,
      "eval_loss": 0.0008142033475451171,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.885828733444214,
      "eval_rewards/margins": 7.454825401306152,
      "eval_rewards/rejected": -4.568996429443359,
      "eval_runtime": 548.1607,
      "eval_samples_per_second": 0.509,
      "eval_steps_per_second": 0.509,
      "step": 250
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.07440608739852905,
      "learning_rate": 5e-07,
      "logits/chosen": -1.7455374002456665,
      "logits/rejected": -1.8116660118103027,
      "logps/chosen": -47.130027770996094,
      "logps/rejected": -80.37494659423828,
      "loss": 0.0005,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.917734384536743,
      "rewards/margins": 8.058551788330078,
      "rewards/rejected": -5.140817642211914,
      "step": 300
    },
    {
      "epoch": 0.26666666666666666,
      "eval_logits/chosen": -1.7649332284927368,
      "eval_logits/rejected": -1.8052014112472534,
      "eval_logps/chosen": -45.18332290649414,
      "eval_logps/rejected": -84.715576171875,
      "eval_loss": 0.0003367140598129481,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.9810502529144287,
      "eval_rewards/margins": 8.570627212524414,
      "eval_rewards/rejected": -5.589577674865723,
      "eval_runtime": 548.0227,
      "eval_samples_per_second": 0.509,
      "eval_steps_per_second": 0.509,
      "step": 300
    },
    {
      "epoch": 0.3111111111111111,
      "grad_norm": 0.23535361886024475,
      "learning_rate": 5e-07,
      "logits/chosen": -1.7623131275177002,
      "logits/rejected": -1.8129371404647827,
      "logps/chosen": -47.72771072387695,
      "logps/rejected": -87.96331787109375,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.999173879623413,
      "rewards/margins": 8.857995986938477,
      "rewards/rejected": -5.858822822570801,
      "step": 350
    },
    {
      "epoch": 0.3111111111111111,
      "eval_logits/chosen": -1.768768310546875,
      "eval_logits/rejected": -1.7995201349258423,
      "eval_logps/chosen": -44.72746276855469,
      "eval_logps/rejected": -90.03182220458984,
      "eval_loss": 0.0002303781802766025,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 3.0266363620758057,
      "eval_rewards/margins": 9.14783763885498,
      "eval_rewards/rejected": -6.121201992034912,
      "eval_runtime": 538.111,
      "eval_samples_per_second": 0.518,
      "eval_steps_per_second": 0.518,
      "step": 350
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 0.012943393550813198,
      "learning_rate": 5e-07,
      "logits/chosen": -1.814980149269104,
      "logits/rejected": -1.8396068811416626,
      "logps/chosen": -45.2337646484375,
      "logps/rejected": -91.96194458007812,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.0036685466766357,
      "rewards/margins": 9.295085906982422,
      "rewards/rejected": -6.291417121887207,
      "step": 400
    },
    {
      "epoch": 0.35555555555555557,
      "eval_logits/chosen": -1.772402048110962,
      "eval_logits/rejected": -1.7961318492889404,
      "eval_logps/chosen": -44.347679138183594,
      "eval_logps/rejected": -93.67013549804688,
      "eval_loss": 0.00018359384557697922,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 3.0646145343780518,
      "eval_rewards/margins": 9.549647331237793,
      "eval_rewards/rejected": -6.485032558441162,
      "eval_runtime": 537.2714,
      "eval_samples_per_second": 0.519,
      "eval_steps_per_second": 0.519,
      "step": 400
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.03291670233011246,
      "learning_rate": 5e-07,
      "logits/chosen": -1.7613160610198975,
      "logits/rejected": -1.7991280555725098,
      "logps/chosen": -45.389404296875,
      "logps/rejected": -94.11991882324219,
      "loss": 0.0005,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.102461814880371,
      "rewards/margins": 9.708904266357422,
      "rewards/rejected": -6.606442451477051,
      "step": 450
    },
    {
      "epoch": 0.4,
      "eval_logits/chosen": -1.7705323696136475,
      "eval_logits/rejected": -1.7885987758636475,
      "eval_logps/chosen": -44.02812194824219,
      "eval_logps/rejected": -96.20520782470703,
      "eval_loss": 0.00015724882541690022,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 3.0965700149536133,
      "eval_rewards/margins": 9.835110664367676,
      "eval_rewards/rejected": -6.738540172576904,
      "eval_runtime": 538.1537,
      "eval_samples_per_second": 0.518,
      "eval_steps_per_second": 0.518,
      "step": 450
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.01980595290660858,
      "learning_rate": 5e-07,
      "logits/chosen": -1.7331243753433228,
      "logits/rejected": -1.7530601024627686,
      "logps/chosen": -46.38658142089844,
      "logps/rejected": -97.57498168945312,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.10678768157959,
      "rewards/margins": 9.987698554992676,
      "rewards/rejected": -6.880909442901611,
      "step": 500
    },
    {
      "epoch": 0.4444444444444444,
      "eval_logits/chosen": -1.7709521055221558,
      "eval_logits/rejected": -1.7846826314926147,
      "eval_logps/chosen": -43.82915496826172,
      "eval_logps/rejected": -98.48301696777344,
      "eval_loss": 0.0001402325287926942,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 3.116466999053955,
      "eval_rewards/margins": 10.08278751373291,
      "eval_rewards/rejected": -6.9663214683532715,
      "eval_runtime": 537.9536,
      "eval_samples_per_second": 0.519,
      "eval_steps_per_second": 0.519,
      "step": 500
    },
    {
      "epoch": 0.4888888888888889,
      "grad_norm": 0.027053946629166603,
      "learning_rate": 5e-07,
      "logits/chosen": -1.7708313465118408,
      "logits/rejected": -1.7873836755752563,
      "logps/chosen": -44.97260665893555,
      "logps/rejected": -99.28631591796875,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.094040870666504,
      "rewards/margins": 10.209158897399902,
      "rewards/rejected": -7.115118026733398,
      "step": 550
    },
    {
      "epoch": 0.4888888888888889,
      "eval_logits/chosen": -1.7708543539047241,
      "eval_logits/rejected": -1.780946969985962,
      "eval_logps/chosen": -43.67860412597656,
      "eval_logps/rejected": -100.63538360595703,
      "eval_loss": 0.0001275104150408879,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 3.1315219402313232,
      "eval_rewards/margins": 10.313079833984375,
      "eval_rewards/rejected": -7.181557655334473,
      "eval_runtime": 536.8647,
      "eval_samples_per_second": 0.52,
      "eval_steps_per_second": 0.52,
      "step": 550
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.01415593083947897,
      "learning_rate": 5e-07,
      "logits/chosen": -1.7776223421096802,
      "logits/rejected": -1.8188660144805908,
      "logps/chosen": -45.774497985839844,
      "logps/rejected": -101.59158325195312,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.103877544403076,
      "rewards/margins": 10.393211364746094,
      "rewards/rejected": -7.289334774017334,
      "step": 600
    },
    {
      "epoch": 0.5333333333333333,
      "eval_logits/chosen": -1.7701154947280884,
      "eval_logits/rejected": -1.7771750688552856,
      "eval_logps/chosen": -43.5441780090332,
      "eval_logps/rejected": -102.60520935058594,
      "eval_loss": 0.00011848945723613724,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 3.1449646949768066,
      "eval_rewards/margins": 10.523506164550781,
      "eval_rewards/rejected": -7.378541469573975,
      "eval_runtime": 537.1558,
      "eval_samples_per_second": 0.519,
      "eval_steps_per_second": 0.519,
      "step": 600
    }
  ],
  "logging_steps": 50,
  "max_steps": 1125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}