{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 239,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0041841004184100415,
      "grad_norm": 2.335511797805867,
      "learning_rate": 2.083333333333333e-08,
      "logits/chosen": -1.168128490447998,
      "logits/rejected": -1.1001684665679932,
      "logps/chosen": -345.7468566894531,
      "logps/rejected": -388.03436279296875,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.04184100418410042,
      "grad_norm": 3.0538125395530042,
      "learning_rate": 2.0833333333333333e-07,
      "logits/chosen": -1.1523456573486328,
      "logits/rejected": -1.216726541519165,
      "logps/chosen": -327.39593505859375,
      "logps/rejected": -330.5606689453125,
      "loss": 0.6932,
      "rewards/accuracies": 0.4583333432674408,
      "rewards/chosen": 0.00016736745601519942,
      "rewards/margins": 0.00036640118923969567,
      "rewards/rejected": -0.00019903367501683533,
      "step": 10
    },
    {
      "epoch": 0.08368200836820083,
      "grad_norm": 2.051052061032609,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -1.1201739311218262,
      "logits/rejected": -1.1966865062713623,
      "logps/chosen": -337.3428649902344,
      "logps/rejected": -310.3174743652344,
      "loss": 0.6931,
      "rewards/accuracies": 0.4906249940395355,
      "rewards/chosen": 0.0008397102355957031,
      "rewards/margins": -0.00019551152945496142,
      "rewards/rejected": 0.001035221852362156,
      "step": 20
    },
    {
      "epoch": 0.12552301255230125,
      "grad_norm": 2.051405044620473,
      "learning_rate": 4.990398100856366e-07,
      "logits/chosen": -1.1592100858688354,
      "logits/rejected": -1.1792274713516235,
      "logps/chosen": -327.45831298828125,
      "logps/rejected": -347.738037109375,
      "loss": 0.6922,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": 0.004632706753909588,
      "rewards/margins": 0.001139953499659896,
      "rewards/rejected": 0.0034927534870803356,
      "step": 30
    },
    {
      "epoch": 0.16736401673640167,
      "grad_norm": 2.280425541026252,
      "learning_rate": 4.931986719649298e-07,
      "logits/chosen": -1.2329559326171875,
      "logits/rejected": -1.273472547531128,
      "logps/chosen": -352.55316162109375,
      "logps/rejected": -314.9006652832031,
      "loss": 0.6902,
      "rewards/accuracies": 0.6343749761581421,
      "rewards/chosen": 0.013729018159210682,
      "rewards/margins": 0.00675433874130249,
      "rewards/rejected": 0.006974679883569479,
      "step": 40
    },
    {
      "epoch": 0.20920502092050208,
      "grad_norm": 1.953444430114674,
      "learning_rate": 4.821741763807186e-07,
      "logits/chosen": -1.2172247171401978,
      "logits/rejected": -1.2943127155303955,
      "logps/chosen": -316.37371826171875,
      "logps/rejected": -312.96661376953125,
      "loss": 0.6874,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": 0.020605359226465225,
      "rewards/margins": 0.01330882590264082,
      "rewards/rejected": 0.007296533323824406,
      "step": 50
    },
    {
      "epoch": 0.2510460251046025,
      "grad_norm": 2.228539317195908,
      "learning_rate": 4.662012913161997e-07,
      "logits/chosen": -1.2346419095993042,
      "logits/rejected": -1.3438832759857178,
      "logps/chosen": -348.8750915527344,
      "logps/rejected": -317.3927307128906,
      "loss": 0.6851,
      "rewards/accuracies": 0.621874988079071,
      "rewards/chosen": 0.03193451464176178,
      "rewards/margins": 0.016508478671312332,
      "rewards/rejected": 0.01542603224515915,
      "step": 60
    },
    {
      "epoch": 0.2928870292887029,
      "grad_norm": 2.1140708636197845,
      "learning_rate": 4.456204510851956e-07,
      "logits/chosen": -1.225719928741455,
      "logits/rejected": -1.2811986207962036,
      "logps/chosen": -344.2880554199219,
      "logps/rejected": -338.8671569824219,
      "loss": 0.6818,
      "rewards/accuracies": 0.640625,
      "rewards/chosen": 0.04076681658625603,
      "rewards/margins": 0.02389165200293064,
      "rewards/rejected": 0.016875162720680237,
      "step": 70
    },
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 1.8046487610538895,
      "learning_rate": 4.2087030056579986e-07,
      "logits/chosen": -1.2622222900390625,
      "logits/rejected": -1.3684368133544922,
      "logps/chosen": -328.14263916015625,
      "logps/rejected": -317.73193359375,
      "loss": 0.6804,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": 0.03804489225149155,
      "rewards/margins": 0.022086812183260918,
      "rewards/rejected": 0.01595807820558548,
      "step": 80
    },
    {
      "epoch": 0.37656903765690375,
      "grad_norm": 2.1792714189731988,
      "learning_rate": 3.9247834624635404e-07,
      "logits/chosen": -1.2576462030410767,
      "logits/rejected": -1.3489463329315186,
      "logps/chosen": -303.93096923828125,
      "logps/rejected": -290.1899719238281,
      "loss": 0.6759,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": 0.050195373594760895,
      "rewards/margins": 0.038325682282447815,
      "rewards/rejected": 0.011869693174958229,
      "step": 90
    },
    {
      "epoch": 0.41841004184100417,
      "grad_norm": 2.022836388367397,
      "learning_rate": 3.610497133404795e-07,
      "logits/chosen": -1.284611701965332,
      "logits/rejected": -1.3313934803009033,
      "logps/chosen": -319.56182861328125,
      "logps/rejected": -312.11492919921875,
      "loss": 0.6762,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": 0.05392398685216904,
      "rewards/margins": 0.04721994698047638,
      "rewards/rejected": 0.006704038940370083,
      "step": 100
    },
    {
      "epoch": 0.41841004184100417,
      "eval_logits/chosen": -1.1861072778701782,
      "eval_logits/rejected": -1.2637689113616943,
      "eval_logps/chosen": -333.8717041015625,
      "eval_logps/rejected": -312.8760681152344,
      "eval_loss": 0.6753271818161011,
      "eval_rewards/accuracies": 0.6626983880996704,
      "eval_rewards/chosen": 0.054593343287706375,
      "eval_rewards/margins": 0.042364537715911865,
      "eval_rewards/rejected": 0.01222880370914936,
      "eval_runtime": 91.2196,
      "eval_samples_per_second": 21.925,
      "eval_steps_per_second": 0.691,
      "step": 100
    },
    {
      "epoch": 0.4602510460251046,
      "grad_norm": 1.8857411098055814,
      "learning_rate": 3.272542485937368e-07,
      "logits/chosen": -1.261444330215454,
      "logits/rejected": -1.3659017086029053,
      "logps/chosen": -330.69903564453125,
      "logps/rejected": -304.2332458496094,
      "loss": 0.6739,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": 0.0468447282910347,
      "rewards/margins": 0.04751918837428093,
      "rewards/rejected": -0.0006744601996615529,
      "step": 110
    },
    {
      "epoch": 0.502092050209205,
      "grad_norm": 2.320788210793853,
      "learning_rate": 2.9181224366319943e-07,
      "logits/chosen": -1.2688654661178589,
      "logits/rejected": -1.3098180294036865,
      "logps/chosen": -307.9932556152344,
      "logps/rejected": -306.85675048828125,
      "loss": 0.6702,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": 0.05479610711336136,
      "rewards/margins": 0.05313174054026604,
      "rewards/rejected": 0.001664361683651805,
      "step": 120
    },
    {
      "epoch": 0.5439330543933054,
      "grad_norm": 1.8823812053819848,
      "learning_rate": 2.55479083351317e-07,
      "logits/chosen": -1.2392737865447998,
      "logits/rejected": -1.3330614566802979,
      "logps/chosen": -348.48931884765625,
      "logps/rejected": -315.95013427734375,
      "loss": 0.6691,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": 0.06406990438699722,
      "rewards/margins": 0.06992693990468979,
      "rewards/rejected": -0.005857035517692566,
      "step": 130
    },
    {
      "epoch": 0.5857740585774058,
      "grad_norm": 5.44373598859388,
      "learning_rate": 2.19029145890313e-07,
      "logits/chosen": -1.2415839433670044,
      "logits/rejected": -1.3235524892807007,
      "logps/chosen": -326.3031005859375,
      "logps/rejected": -311.26214599609375,
      "loss": 0.6678,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": 0.054818809032440186,
      "rewards/margins": 0.05220777913928032,
      "rewards/rejected": 0.002611032919958234,
      "step": 140
    },
    {
      "epoch": 0.6276150627615062,
      "grad_norm": 1.939512273689077,
      "learning_rate": 1.8323929841460178e-07,
      "logits/chosen": -1.227124571800232,
      "logits/rejected": -1.3289756774902344,
      "logps/chosen": -326.6729736328125,
      "logps/rejected": -308.0332336425781,
      "loss": 0.666,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": 0.035384777933359146,
      "rewards/margins": 0.051828302443027496,
      "rewards/rejected": -0.01644352450966835,
      "step": 150
    },
    {
      "epoch": 0.6694560669456067,
      "grad_norm": 1.9253405708401141,
      "learning_rate": 1.488723393865766e-07,
      "logits/chosen": -1.264524221420288,
      "logits/rejected": -1.307389736175537,
      "logps/chosen": -344.7317199707031,
      "logps/rejected": -312.7427978515625,
      "loss": 0.6612,
      "rewards/accuracies": 0.684374988079071,
      "rewards/chosen": 0.04708511754870415,
      "rewards/margins": 0.07312874495983124,
      "rewards/rejected": -0.02604362927377224,
      "step": 160
    },
    {
      "epoch": 0.7112970711297071,
      "grad_norm": 1.9507035640244035,
      "learning_rate": 1.1666074087171627e-07,
      "logits/chosen": -1.2524497509002686,
      "logits/rejected": -1.2847630977630615,
      "logps/chosen": -332.83990478515625,
      "logps/rejected": -329.17608642578125,
      "loss": 0.6645,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": 0.03406905010342598,
      "rewards/margins": 0.05726125091314316,
      "rewards/rejected": -0.023192204535007477,
      "step": 170
    },
    {
      "epoch": 0.7531380753138075,
      "grad_norm": 1.8449874305732954,
      "learning_rate": 8.729103716819111e-08,
      "logits/chosen": -1.2439481019973755,
      "logits/rejected": -1.2792072296142578,
      "logps/chosen": -332.6515197753906,
      "logps/rejected": -310.3517150878906,
      "loss": 0.6615,
      "rewards/accuracies": 0.6781250238418579,
      "rewards/chosen": 0.04834641143679619,
      "rewards/margins": 0.07643269002437592,
      "rewards/rejected": -0.028086284175515175,
      "step": 180
    },
    {
      "epoch": 0.7949790794979079,
      "grad_norm": 2.2113847121012986,
      "learning_rate": 6.138919252022435e-08,
      "logits/chosen": -1.270702600479126,
      "logits/rejected": -1.3235771656036377,
      "logps/chosen": -313.1623229980469,
      "logps/rejected": -310.3594665527344,
      "loss": 0.6592,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": 0.020944807678461075,
      "rewards/margins": 0.0681905671954155,
      "rewards/rejected": -0.04724575951695442,
      "step": 190
    },
    {
      "epoch": 0.8368200836820083,
      "grad_norm": 1.9219438856219404,
      "learning_rate": 3.9507259776993954e-08,
      "logits/chosen": -1.2548792362213135,
      "logits/rejected": -1.2600829601287842,
      "logps/chosen": -339.12457275390625,
      "logps/rejected": -337.1640625,
      "loss": 0.6604,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": 0.040566906332969666,
      "rewards/margins": 0.07623802870512009,
      "rewards/rejected": -0.03567112237215042,
      "step": 200
    },
    {
      "epoch": 0.8368200836820083,
      "eval_logits/chosen": -1.1793808937072754,
      "eval_logits/rejected": -1.2523409128189087,
      "eval_logps/chosen": -335.1041564941406,
      "eval_logps/rejected": -317.20819091796875,
      "eval_loss": 0.6639689803123474,
      "eval_rewards/accuracies": 0.670634925365448,
      "eval_rewards/chosen": 0.04226858913898468,
      "eval_rewards/margins": 0.07336094230413437,
      "eval_rewards/rejected": -0.031092360615730286,
      "eval_runtime": 91.7242,
      "eval_samples_per_second": 21.804,
      "eval_steps_per_second": 0.687,
      "step": 200
    },
    {
      "epoch": 0.8786610878661087,
      "grad_norm": 2.0461724170667646,
      "learning_rate": 2.2111614344599684e-08,
      "logits/chosen": -1.1933133602142334,
      "logits/rejected": -1.3109954595565796,
      "logps/chosen": -347.81976318359375,
      "logps/rejected": -335.7594909667969,
      "loss": 0.6613,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": 0.03740222379565239,
      "rewards/margins": 0.05939198657870293,
      "rewards/rejected": -0.021989766508340836,
      "step": 210
    },
    {
      "epoch": 0.9205020920502092,
      "grad_norm": 3.392645562933325,
      "learning_rate": 9.57301420397924e-09,
      "logits/chosen": -1.1992428302764893,
      "logits/rejected": -1.243272066116333,
      "logps/chosen": -338.14166259765625,
      "logps/rejected": -323.56817626953125,
      "loss": 0.6585,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": 0.04126737266778946,
      "rewards/margins": 0.07338305562734604,
      "rewards/rejected": -0.032115690410137177,
      "step": 220
    },
    {
      "epoch": 0.9623430962343096,
      "grad_norm": 1.9082006258080877,
      "learning_rate": 2.158697848236607e-09,
      "logits/chosen": -1.2281768321990967,
      "logits/rejected": -1.2996705770492554,
      "logps/chosen": -334.5081787109375,
      "logps/rejected": -336.0547790527344,
      "loss": 0.6611,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": 0.029226357117295265,
      "rewards/margins": 0.054667878895998,
      "rewards/rejected": -0.025441521778702736,
      "step": 230
    },
    {
      "epoch": 1.0,
      "step": 239,
      "total_flos": 0.0,
      "train_loss": 0.6732059223382543,
      "train_runtime": 6951.1203,
      "train_samples_per_second": 8.795,
      "train_steps_per_second": 0.034
    }
  ],
  "logging_steps": 10,
  "max_steps": 239,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}