{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9221902017291066,
  "eval_steps": 500,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09221902017291066,
      "grad_norm": 0.3422877788543701,
      "learning_rate": 5e-07,
      "logits/chosen": 0.1349429488182068,
      "logits/rejected": 0.2280968725681305,
      "logps/chosen": -136.6981658935547,
      "logps/ref_chosen": -126.77554321289062,
      "logps/ref_rejected": -155.40750122070312,
      "logps/rejected": -165.48233032226562,
      "loss": 0.5773,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.09922622889280319,
      "rewards/margins": 0.001521922298707068,
      "rewards/rejected": -0.10074815154075623,
      "sft_loss": 0.46219536662101746,
      "step": 1
    },
    {
      "epoch": 0.1844380403458213,
      "grad_norm": 0.3397272229194641,
      "learning_rate": 4.849231551964771e-07,
      "logits/chosen": 0.10780894756317139,
      "logits/rejected": 0.2713747024536133,
      "logps/chosen": -130.30294799804688,
      "logps/ref_chosen": -119.76988220214844,
      "logps/ref_rejected": -154.81539916992188,
      "logps/rejected": -165.45050048828125,
      "loss": 0.5844,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": -0.1053306981921196,
      "rewards/margins": 0.0010203586425632238,
      "rewards/rejected": -0.10635105520486832,
      "sft_loss": 0.476193368434906,
      "step": 2
    },
    {
      "epoch": 0.276657060518732,
      "grad_norm": 0.3958917260169983,
      "learning_rate": 4.415111107797445e-07,
      "logits/chosen": 0.08285358548164368,
      "logits/rejected": 0.22937394678592682,
      "logps/chosen": -130.75489807128906,
      "logps/ref_chosen": -121.35408782958984,
      "logps/ref_rejected": -153.38841247558594,
      "logps/rejected": -162.9373016357422,
      "loss": 0.5853,
      "rewards/accuracies": 0.6328125,
      "rewards/chosen": -0.09400798380374908,
      "rewards/margins": 0.0014810654101893306,
      "rewards/rejected": -0.09548905491828918,
      "sft_loss": 0.47810834646224976,
      "step": 3
    },
    {
      "epoch": 0.3688760806916426,
      "grad_norm": 0.36922165751457214,
      "learning_rate": 3.75e-07,
      "logits/chosen": 0.1300196349620819,
      "logits/rejected": 0.24173876643180847,
      "logps/chosen": -130.75794982910156,
      "logps/ref_chosen": -121.80622100830078,
      "logps/ref_rejected": -147.26390075683594,
      "logps/rejected": -156.2241973876953,
      "loss": 0.581,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": -0.08951741456985474,
      "rewards/margins": 8.561517461203039e-05,
      "rewards/rejected": -0.08960302919149399,
      "sft_loss": 0.4689139723777771,
      "step": 4
    },
    {
      "epoch": 0.4610951008645533,
      "grad_norm": 0.39064234495162964,
      "learning_rate": 2.934120444167326e-07,
      "logits/chosen": 0.13762706518173218,
      "logits/rejected": 0.2819656431674957,
      "logps/chosen": -129.11767578125,
      "logps/ref_chosen": -121.6624984741211,
      "logps/ref_rejected": -154.23062133789062,
      "logps/rejected": -161.6747283935547,
      "loss": 0.5752,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.0745517909526825,
      "rewards/margins": -0.00011071082553826272,
      "rewards/rejected": -0.0744410902261734,
      "sft_loss": 0.4572478234767914,
      "step": 5
    },
    {
      "epoch": 0.553314121037464,
      "grad_norm": 0.5450302362442017,
      "learning_rate": 2.065879555832674e-07,
      "logits/chosen": 0.12577247619628906,
      "logits/rejected": 0.23831848800182343,
      "logps/chosen": -130.1405487060547,
      "logps/ref_chosen": -123.26316833496094,
      "logps/ref_rejected": -151.7885284423828,
      "logps/rejected": -158.7672119140625,
      "loss": 0.5738,
      "rewards/accuracies": 0.6171875,
      "rewards/chosen": -0.06877387315034866,
      "rewards/margins": 0.0010129868751391768,
      "rewards/rejected": -0.06978685408830643,
      "sft_loss": 0.45494452118873596,
      "step": 6
    },
    {
      "epoch": 0.6455331412103746,
      "grad_norm": 0.34464988112449646,
      "learning_rate": 1.2500000000000005e-07,
      "logits/chosen": 0.13395586609840393,
      "logits/rejected": 0.2669757306575775,
      "logps/chosen": -127.01806640625,
      "logps/ref_chosen": -123.50520324707031,
      "logps/ref_rejected": -162.37644958496094,
      "logps/rejected": -165.93678283691406,
      "loss": 0.569,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.03512854874134064,
      "rewards/margins": 0.0004748337669298053,
      "rewards/rejected": -0.03560338541865349,
      "sft_loss": 0.44512274861335754,
      "step": 7
    },
    {
      "epoch": 0.7377521613832853,
      "grad_norm": 0.34306466579437256,
      "learning_rate": 5.848888922025552e-08,
      "logits/chosen": 0.0844278633594513,
      "logits/rejected": 0.1925371140241623,
      "logps/chosen": -122.23373413085938,
      "logps/ref_chosen": -120.31718444824219,
      "logps/ref_rejected": -149.6009979248047,
      "logps/rejected": -151.5816192626953,
      "loss": 0.5717,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.01916549727320671,
      "rewards/margins": 0.0006407537148334086,
      "rewards/rejected": -0.019806252792477608,
      "sft_loss": 0.45050206780433655,
      "step": 8
    },
    {
      "epoch": 0.829971181556196,
      "grad_norm": 0.40833377838134766,
      "learning_rate": 1.507684480352292e-08,
      "logits/chosen": 0.11135825514793396,
      "logits/rejected": 0.2538539171218872,
      "logps/chosen": -125.90987396240234,
      "logps/ref_chosen": -124.37432861328125,
      "logps/ref_rejected": -150.76690673828125,
      "logps/rejected": -152.4391632080078,
      "loss": 0.5723,
      "rewards/accuracies": 0.5703125,
      "rewards/chosen": -0.015355447307229042,
      "rewards/margins": 0.0013672513887286186,
      "rewards/rejected": -0.016722697764635086,
      "sft_loss": 0.45217180252075195,
      "step": 9
    },
    {
      "epoch": 0.9221902017291066,
      "grad_norm": 0.30012214183807373,
      "learning_rate": 0.0,
      "logits/chosen": 0.1120913028717041,
      "logits/rejected": 0.30093058943748474,
      "logps/chosen": -123.66395568847656,
      "logps/ref_chosen": -122.19993591308594,
      "logps/ref_rejected": -156.10472106933594,
      "logps/rejected": -157.66062927246094,
      "loss": 0.569,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": -0.01464010588824749,
      "rewards/margins": 0.0009190225973725319,
      "rewards/rejected": -0.015559129416942596,
      "sft_loss": 0.4453127384185791,
      "step": 10
    },
    {
      "epoch": 0.9221902017291066,
      "step": 10,
      "total_flos": 0.0,
      "train_loss": 0.5759013295173645,
      "train_runtime": 619.8336,
      "train_samples_per_second": 2.238,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}