{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6872160934458145,
  "eval_steps": 25,
  "global_step": 325,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12978585334198572,
      "grad_norm": 64.38245391845703,
      "learning_rate": 5e-07,
      "logits/chosen": -1.9029332399368286,
      "logits/rejected": -2.053706407546997,
      "logps/chosen": -40.4902458190918,
      "logps/rejected": -25.3558406829834,
      "loss": 0.5968,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": 0.17978771030902863,
      "rewards/margins": 0.2116149365901947,
      "rewards/rejected": -0.031827230006456375,
      "step": 25
    },
    {
      "epoch": 0.12978585334198572,
      "eval_logits/chosen": -1.9241564273834229,
      "eval_logits/rejected": -2.0354554653167725,
      "eval_logps/chosen": -38.01565933227539,
      "eval_logps/rejected": -26.247798919677734,
      "eval_loss": 0.454921156167984,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 0.4693543612957001,
      "eval_rewards/margins": 0.5530296564102173,
      "eval_rewards/rejected": -0.083675317466259,
      "eval_runtime": 140.2735,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 25
    },
    {
      "epoch": 0.25957170668397145,
      "grad_norm": 40.388370513916016,
      "learning_rate": 5e-07,
      "logits/chosen": -1.9818081855773926,
      "logits/rejected": -2.0913336277008057,
      "logps/chosen": -34.876155853271484,
      "logps/rejected": -27.348379135131836,
      "loss": 0.3498,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.7389992475509644,
      "rewards/margins": 0.8872407674789429,
      "rewards/rejected": -0.14824140071868896,
      "step": 50
    },
    {
      "epoch": 0.25957170668397145,
      "eval_logits/chosen": -2.0032734870910645,
      "eval_logits/rejected": -2.0942301750183105,
      "eval_logps/chosen": -32.67618942260742,
      "eval_logps/rejected": -27.68032455444336,
      "eval_loss": 0.2584493160247803,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.003301739692688,
      "eval_rewards/margins": 1.2302298545837402,
      "eval_rewards/rejected": -0.22692809998989105,
      "eval_runtime": 140.4996,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 50
    },
    {
      "epoch": 0.3893575600259572,
      "grad_norm": 29.64296531677246,
      "learning_rate": 5e-07,
      "logits/chosen": -2.0457983016967773,
      "logits/rejected": -2.112644910812378,
      "logps/chosen": -29.919565200805664,
      "logps/rejected": -29.504945755004883,
      "loss": 0.1989,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.2103655338287354,
      "rewards/margins": 1.545750379562378,
      "rewards/rejected": -0.33538496494293213,
      "step": 75
    },
    {
      "epoch": 0.3893575600259572,
      "eval_logits/chosen": -2.056947946548462,
      "eval_logits/rejected": -2.1299691200256348,
      "eval_logps/chosen": -28.753862380981445,
      "eval_logps/rejected": -30.1635684967041,
      "eval_loss": 0.14609307050704956,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.3955342769622803,
      "eval_rewards/margins": 1.8707866668701172,
      "eval_rewards/rejected": -0.4752523899078369,
      "eval_runtime": 140.5032,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 75
    },
    {
      "epoch": 0.5191434133679429,
      "grad_norm": 15.144841194152832,
      "learning_rate": 5e-07,
      "logits/chosen": -1.9773192405700684,
      "logits/rejected": -2.081099271774292,
      "logps/chosen": -27.414873123168945,
      "logps/rejected": -32.0408935546875,
      "loss": 0.1067,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5709582567214966,
      "rewards/margins": 2.2332875728607178,
      "rewards/rejected": -0.6623291969299316,
      "step": 100
    },
    {
      "epoch": 0.5191434133679429,
      "eval_logits/chosen": -2.0898687839508057,
      "eval_logits/rejected": -2.1492760181427,
      "eval_logps/chosen": -25.59548568725586,
      "eval_logps/rejected": -34.3668098449707,
      "eval_loss": 0.07369648665189743,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.7113721370697021,
      "eval_rewards/margins": 2.6069483757019043,
      "eval_rewards/rejected": -0.8955763578414917,
      "eval_runtime": 140.7464,
      "eval_samples_per_second": 0.583,
      "eval_steps_per_second": 0.583,
      "step": 100
    },
    {
      "epoch": 0.6489292667099286,
      "grad_norm": 6.891702175140381,
      "learning_rate": 5e-07,
      "logits/chosen": -2.0497264862060547,
      "logits/rejected": -2.113835334777832,
      "logps/chosen": -25.3034725189209,
      "logps/rejected": -37.589698791503906,
      "loss": 0.0501,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.810562014579773,
      "rewards/margins": 3.047495126724243,
      "rewards/rejected": -1.2369328737258911,
      "step": 125
    },
    {
      "epoch": 0.6489292667099286,
      "eval_logits/chosen": -2.117217540740967,
      "eval_logits/rejected": -2.166142463684082,
      "eval_logps/chosen": -24.0587100982666,
      "eval_logps/rejected": -42.671173095703125,
      "eval_loss": 0.02941078133881092,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.8650493621826172,
      "eval_rewards/margins": 3.59106183052063,
      "eval_rewards/rejected": -1.7260124683380127,
      "eval_runtime": 140.3665,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 125
    },
    {
      "epoch": 0.7787151200519143,
      "grad_norm": 2.0042898654937744,
      "learning_rate": 5e-07,
      "logits/chosen": -2.1526224613189697,
      "logits/rejected": -2.214064598083496,
      "logps/chosen": -24.000539779663086,
      "logps/rejected": -47.32194900512695,
      "loss": 0.0177,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8812981843948364,
      "rewards/margins": 4.163049697875977,
      "rewards/rejected": -2.2817516326904297,
      "step": 150
    },
    {
      "epoch": 0.7787151200519143,
      "eval_logits/chosen": -2.1497561931610107,
      "eval_logits/rejected": -2.1853065490722656,
      "eval_logps/chosen": -23.404022216796875,
      "eval_logps/rejected": -53.90214538574219,
      "eval_loss": 0.009417637251317501,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.930518388748169,
      "eval_rewards/margins": 4.779628276824951,
      "eval_rewards/rejected": -2.849109649658203,
      "eval_runtime": 140.3986,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 150
    },
    {
      "epoch": 0.9085009733939,
      "grad_norm": 0.955883264541626,
      "learning_rate": 5e-07,
      "logits/chosen": -2.123363971710205,
      "logits/rejected": -2.164508581161499,
      "logps/chosen": -23.00540542602539,
      "logps/rejected": -57.601531982421875,
      "loss": 0.0062,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9459924697875977,
      "rewards/margins": 5.212995529174805,
      "rewards/rejected": -3.2670037746429443,
      "step": 175
    },
    {
      "epoch": 0.9085009733939,
      "eval_logits/chosen": -2.178921937942505,
      "eval_logits/rejected": -2.198212146759033,
      "eval_logps/chosen": -22.801753997802734,
      "eval_logps/rejected": -63.14146423339844,
      "eval_loss": 0.0035098514053970575,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.990744948387146,
      "eval_rewards/margins": 5.763787269592285,
      "eval_rewards/rejected": -3.7730419635772705,
      "eval_runtime": 140.3451,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 175
    },
    {
      "epoch": 1.0382868267358858,
      "grad_norm": 0.22394584119319916,
      "learning_rate": 5e-07,
      "logits/chosen": -2.161266803741455,
      "logits/rejected": -2.1958141326904297,
      "logps/chosen": -22.247957229614258,
      "logps/rejected": -67.36149597167969,
      "loss": 0.0024,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9859832525253296,
      "rewards/margins": 6.165727615356445,
      "rewards/rejected": -4.179744720458984,
      "step": 200
    },
    {
      "epoch": 1.0382868267358858,
      "eval_logits/chosen": -2.2002382278442383,
      "eval_logits/rejected": -2.2000482082366943,
      "eval_logps/chosen": -22.139089584350586,
      "eval_logps/rejected": -72.42859649658203,
      "eval_loss": 0.001294078305363655,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.057011604309082,
      "eval_rewards/margins": 6.7587666511535645,
      "eval_rewards/rejected": -4.701755523681641,
      "eval_runtime": 140.2609,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 200
    },
    {
      "epoch": 1.1680726800778716,
      "grad_norm": 0.10501434653997421,
      "learning_rate": 5e-07,
      "logits/chosen": -2.195300340652466,
      "logits/rejected": -2.1965079307556152,
      "logps/chosen": -22.331872940063477,
      "logps/rejected": -75.86756896972656,
      "loss": 0.0009,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.0966131687164307,
      "rewards/margins": 7.172005653381348,
      "rewards/rejected": -5.075392723083496,
      "step": 225
    },
    {
      "epoch": 1.1680726800778716,
      "eval_logits/chosen": -2.2084293365478516,
      "eval_logits/rejected": -2.1937882900238037,
      "eval_logps/chosen": -21.62232208251953,
      "eval_logps/rejected": -80.40409088134766,
      "eval_loss": 0.0005517782410606742,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.1086885929107666,
      "eval_rewards/margins": 7.607994079589844,
      "eval_rewards/rejected": -5.499305725097656,
      "eval_runtime": 140.4282,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 225
    },
    {
      "epoch": 1.2978585334198574,
      "grad_norm": 0.12079267203807831,
      "learning_rate": 5e-07,
      "logits/chosen": -2.2555391788482666,
      "logits/rejected": -2.2508366107940674,
      "logps/chosen": -21.726062774658203,
      "logps/rejected": -83.389892578125,
      "loss": 0.0005,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.083451271057129,
      "rewards/margins": 7.826067924499512,
      "rewards/rejected": -5.742616653442383,
      "step": 250
    },
    {
      "epoch": 1.2978585334198574,
      "eval_logits/chosen": -2.2094874382019043,
      "eval_logits/rejected": -2.183870553970337,
      "eval_logps/chosen": -21.337453842163086,
      "eval_logps/rejected": -85.59798431396484,
      "eval_loss": 0.00031946305534802377,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.1371753215789795,
      "eval_rewards/margins": 8.155868530273438,
      "eval_rewards/rejected": -6.018693923950195,
      "eval_runtime": 140.1924,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 250
    },
    {
      "epoch": 1.427644386761843,
      "grad_norm": 0.03313301131129265,
      "learning_rate": 5e-07,
      "logits/chosen": -2.1494364738464355,
      "logits/rejected": -2.1585676670074463,
      "logps/chosen": -21.17609405517578,
      "logps/rejected": -86.35639953613281,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.207101821899414,
      "rewards/margins": 8.373501777648926,
      "rewards/rejected": -6.166399955749512,
      "step": 275
    },
    {
      "epoch": 1.427644386761843,
      "eval_logits/chosen": -2.2073066234588623,
      "eval_logits/rejected": -2.174794912338257,
      "eval_logps/chosen": -21.185022354125977,
      "eval_logps/rejected": -89.19384765625,
      "eval_loss": 0.00021948952053207904,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.1524181365966797,
      "eval_rewards/margins": 8.530698776245117,
      "eval_rewards/rejected": -6.378280162811279,
      "eval_runtime": 140.1155,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 275
    },
    {
      "epoch": 1.5574302401038287,
      "grad_norm": 0.03680599108338356,
      "learning_rate": 5e-07,
      "logits/chosen": -2.191593647003174,
      "logits/rejected": -2.178603410720825,
      "logps/chosen": -20.61453628540039,
      "logps/rejected": -89.9397201538086,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.125040292739868,
      "rewards/margins": 8.604403495788574,
      "rewards/rejected": -6.479363918304443,
      "step": 300
    },
    {
      "epoch": 1.5574302401038287,
      "eval_logits/chosen": -2.2045023441314697,
      "eval_logits/rejected": -2.1671271324157715,
      "eval_logps/chosen": -21.077783584594727,
      "eval_logps/rejected": -91.99667358398438,
      "eval_loss": 0.0001641342241782695,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.163142204284668,
      "eval_rewards/margins": 8.821704864501953,
      "eval_rewards/rejected": -6.658563613891602,
      "eval_runtime": 140.5186,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 300
    },
    {
      "epoch": 1.6872160934458145,
      "grad_norm": 0.028771178796887398,
      "learning_rate": 5e-07,
      "logits/chosen": -2.203645706176758,
      "logits/rejected": -2.166923761367798,
      "logps/chosen": -20.792112350463867,
      "logps/rejected": -93.08446502685547,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.144740104675293,
      "rewards/margins": 8.90222454071045,
      "rewards/rejected": -6.7574849128723145,
      "step": 325
    },
    {
      "epoch": 1.6872160934458145,
      "eval_logits/chosen": -2.2036421298980713,
      "eval_logits/rejected": -2.1623051166534424,
      "eval_logps/chosen": -20.99409294128418,
      "eval_logps/rejected": -94.2848892211914,
      "eval_loss": 0.000129787833429873,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.171510934829712,
      "eval_rewards/margins": 9.0588960647583,
      "eval_rewards/rejected": -6.887384414672852,
      "eval_runtime": 140.302,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 325
    }
  ],
  "logging_steps": 25,
  "max_steps": 384,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 25,
  "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
}