{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1095,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004571428571428572,
      "grad_norm": 8.67539865837776,
      "learning_rate": 0.0,
      "logps_logqs/chosen": -0.0693359375,
      "logqs/chosen": -2.375,
      "logqs/rejected": -2.375,
      "logqs_logps/rejected": 0.0693359375,
      "loss": 0.6643,
      "step": 1
    },
    {
      "epoch": 0.22857142857142856,
      "grad_norm": 0.14019721630763493,
      "learning_rate": 8.90909090909091e-05,
      "logps_logqs/chosen": 0.018798828125,
      "logqs/chosen": -4.15625,
      "logqs/rejected": -4.15625,
      "logqs_logps/rejected": -0.018310546875,
      "loss": 0.4195,
      "step": 50
    },
    {
      "epoch": 0.45714285714285713,
      "grad_norm": 0.45905529641569304,
      "learning_rate": 9.955899960329547e-05,
      "logps_logqs/chosen": 0.0380859375,
      "logqs/chosen": -4.8125,
      "logqs/rejected": -4.78125,
      "logqs_logps/rejected": -0.037841796875,
      "loss": 0.2379,
      "step": 100
    },
    {
      "epoch": 0.6857142857142857,
      "grad_norm": 0.24812229199411728,
      "learning_rate": 9.799779392165715e-05,
      "logps_logqs/chosen": 0.014404296875,
      "logqs/chosen": -3.703125,
      "logqs/rejected": -3.6875,
      "logqs_logps/rejected": -0.0140380859375,
      "loss": 0.2077,
      "step": 150
    },
    {
      "epoch": 0.9142857142857143,
      "grad_norm": 0.3582955582094429,
      "learning_rate": 9.534371804252728e-05,
      "logps_logqs/chosen": 0.01251220703125,
      "logqs/chosen": -3.578125,
      "logqs/rejected": -3.578125,
      "logqs_logps/rejected": -0.01220703125,
      "loss": 0.1949,
      "step": 200
    },
    {
      "epoch": 1.1417142857142857,
      "grad_norm": 0.3905038311837086,
      "learning_rate": 9.165720308561348e-05,
      "logps_logqs/chosen": 0.0089111328125,
      "logqs/chosen": -3.671875,
      "logqs/rejected": -3.671875,
      "logqs_logps/rejected": -0.0087890625,
      "loss": 0.1899,
      "step": 250
    },
    {
      "epoch": 1.3702857142857143,
      "grad_norm": 0.3292403853943784,
      "learning_rate": 8.702218795726617e-05,
      "logps_logqs/chosen": 0.006988525390625,
      "logqs/chosen": -3.609375,
      "logqs/rejected": -3.609375,
      "logqs_logps/rejected": -0.00689697265625,
      "loss": 0.1779,
      "step": 300
    },
    {
      "epoch": 1.5988571428571428,
      "grad_norm": 0.44189530758031564,
      "learning_rate": 8.154420813078153e-05,
      "logps_logqs/chosen": -1.621246337890625e-05,
      "logqs/chosen": -3.453125,
      "logqs/rejected": -3.453125,
      "logqs_logps/rejected": 0.0002841949462890625,
      "loss": 0.1688,
      "step": 350
    },
    {
      "epoch": 1.8274285714285714,
      "grad_norm": 0.29224407333605207,
      "learning_rate": 7.534799269067953e-05,
      "logps_logqs/chosen": -0.0031890869140625,
      "logqs/chosen": -3.28125,
      "logqs/rejected": -3.296875,
      "logqs_logps/rejected": 0.003509521484375,
      "loss": 0.1546,
      "step": 400
    },
    {
      "epoch": 2.0548571428571427,
      "grad_norm": 0.3845748871005478,
      "learning_rate": 6.857462435427786e-05,
      "logps_logqs/chosen": 0.0002040863037109375,
      "logqs/chosen": -3.4375,
      "logqs/rejected": -3.4375,
      "logqs_logps/rejected": -2.276897430419922e-05,
      "loss": 0.1534,
      "step": 450
    },
    {
      "epoch": 2.2834285714285714,
      "grad_norm": 0.5119536467282476,
      "learning_rate": 6.13783271345295e-05,
      "logps_logqs/chosen": 0.002655029296875,
      "logqs/chosen": -3.453125,
      "logqs/rejected": -3.453125,
      "logqs_logps/rejected": -0.0024871826171875,
      "loss": 0.1458,
      "step": 500
    },
    {
      "epoch": 2.512,
      "grad_norm": 0.3943965132280447,
      "learning_rate": 5.392295478639225e-05,
      "logps_logqs/chosen": 0.005615234375,
      "logqs/chosen": -3.59375,
      "logqs/rejected": -3.578125,
      "logqs_logps/rejected": -0.005340576171875,
      "loss": 0.1452,
      "step": 550
    },
    {
      "epoch": 2.7405714285714287,
      "grad_norm": 0.3862365771403141,
      "learning_rate": 4.6378259991911886e-05,
      "logps_logqs/chosen": 0.005889892578125,
      "logqs/chosen": -3.65625,
      "logqs/rejected": -3.65625,
      "logqs_logps/rejected": -0.00579833984375,
      "loss": 0.1436,
      "step": 600
    },
    {
      "epoch": 2.9691428571428573,
      "grad_norm": 0.27240909322764106,
      "learning_rate": 3.891602923159865e-05,
      "logps_logqs/chosen": 0.0068359375,
      "logqs/chosen": -3.5,
      "logqs/rejected": -3.5,
      "logqs_logps/rejected": -0.0064697265625,
      "loss": 0.1392,
      "step": 650
    },
    {
      "epoch": 3.1965714285714286,
      "grad_norm": 0.4575552509639352,
      "learning_rate": 3.170617134788939e-05,
      "logps_logqs/chosen": 0.00909423828125,
      "logqs/chosen": -3.71875,
      "logqs/rejected": -3.71875,
      "logqs_logps/rejected": -0.00897216796875,
      "loss": 0.1351,
      "step": 700
    },
    {
      "epoch": 3.4251428571428573,
      "grad_norm": 0.4328511479525827,
      "learning_rate": 2.49128488608807e-05,
      "logps_logqs/chosen": 0.0111083984375,
      "logqs/chosen": -3.6875,
      "logqs/rejected": -3.6875,
      "logqs_logps/rejected": -0.01104736328125,
      "loss": 0.1318,
      "step": 750
    },
    {
      "epoch": 3.653714285714286,
      "grad_norm": 0.6170795996671066,
      "learning_rate": 1.8690740123084316e-05,
      "logps_logqs/chosen": 0.01092529296875,
      "logqs/chosen": -3.796875,
      "logqs/rejected": -3.796875,
      "logqs_logps/rejected": -0.0107421875,
      "loss": 0.1335,
      "step": 800
    },
    {
      "epoch": 3.8822857142857146,
      "grad_norm": 0.38700731830789775,
      "learning_rate": 1.3181517420859696e-05,
      "logps_logqs/chosen": 0.01104736328125,
      "logqs/chosen": -3.796875,
      "logqs/rejected": -3.796875,
      "logqs_logps/rejected": -0.01104736328125,
      "loss": 0.1316,
      "step": 850
    },
    {
      "epoch": 4.109714285714285,
      "grad_norm": 0.30337341707201704,
      "learning_rate": 8.510621213250247e-06,
      "logps_logqs/chosen": 0.01214599609375,
      "logqs/chosen": -3.75,
      "logqs/rejected": -3.75,
      "logqs_logps/rejected": -0.01202392578125,
      "loss": 0.1259,
      "step": 900
    },
    {
      "epoch": 4.338285714285714,
      "grad_norm": 0.28556277722070145,
      "learning_rate": 4.7844039561460385e-06,
      "logps_logqs/chosen": 0.01361083984375,
      "logqs/chosen": -3.84375,
      "logqs/rejected": -3.828125,
      "logqs_logps/rejected": -0.0135498046875,
      "loss": 0.1276,
      "step": 950
    },
    {
      "epoch": 4.566857142857143,
      "grad_norm": 0.3198528007608756,
      "learning_rate": 2.087708544541689e-06,
      "logps_logqs/chosen": 0.015380859375,
      "logqs/chosen": -3.671875,
      "logqs/rejected": -3.671875,
      "logqs_logps/rejected": -0.01519775390625,
      "loss": 0.1246,
      "step": 1000
    },
    {
      "epoch": 4.795428571428571,
      "grad_norm": 0.47113628100481,
      "learning_rate": 4.819365097630813e-07,
      "logps_logqs/chosen": 0.0146484375,
      "logqs/chosen": -3.859375,
      "logqs/rejected": -3.859375,
      "logqs_logps/rejected": -0.01458740234375,
      "loss": 0.1283,
      "step": 1050
    },
    {
      "epoch": 5.0,
      "step": 1095,
      "total_flos": 0.0,
      "train_loss": 0.16590411390888093,
      "train_runtime": 31619.711,
      "train_samples_per_second": 1.107,
      "train_steps_per_second": 0.035
    }
  ],
  "logging_steps": 50,
  "max_steps": 1095,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "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": 1,
  "trial_name": null,
  "trial_params": null
}