{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 34.285714285714285,
  "eval_steps": 10,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 0,
      "loss": 1.3795,
      "step": 1
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 2.172917366027832,
      "learning_rate": 0.0,
      "loss": 0.2103,
      "step": 2
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.172917366027832,
      "learning_rate": 0.0,
      "loss": 0.9608,
      "step": 3
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 0.879162609577179,
      "learning_rate": 0.0002,
      "loss": 0.2294,
      "step": 4
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.879162609577179,
      "learning_rate": 0.0002,
      "loss": 0.8781,
      "step": 5
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 0.752986490726471,
      "learning_rate": 0.0002,
      "loss": 0.5153,
      "step": 6
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.752986490726471,
      "learning_rate": 0.0002,
      "loss": 0.5011,
      "step": 7
    },
    {
      "epoch": 4.571428571428571,
      "grad_norm": 0.5658770203590393,
      "learning_rate": 0.0002,
      "loss": 0.5315,
      "step": 8
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5658770203590393,
      "learning_rate": 0.0002,
      "loss": 0.3573,
      "step": 9
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 0.6311860084533691,
      "learning_rate": 0.0002,
      "loss": 0.632,
      "step": 10
    },
    {
      "epoch": 5.714285714285714,
      "eval_loss": 0.794529914855957,
      "eval_runtime": 1.124,
      "eval_samples_per_second": 6.228,
      "eval_steps_per_second": 6.228,
      "step": 10
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6311860084533691,
      "learning_rate": 0.0002,
      "loss": 0.1586,
      "step": 11
    },
    {
      "epoch": 6.857142857142857,
      "grad_norm": 0.6447564959526062,
      "learning_rate": 0.0002,
      "loss": 0.5467,
      "step": 12
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.6447564959526062,
      "learning_rate": 0.0002,
      "loss": 0.1358,
      "step": 13
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.502592921257019,
      "learning_rate": 0.0002,
      "loss": 0.586,
      "step": 14
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.502592921257019,
      "learning_rate": 0.0002,
      "loss": 0.4717,
      "step": 15
    },
    {
      "epoch": 9.142857142857142,
      "grad_norm": 0.515450119972229,
      "learning_rate": 0.0002,
      "loss": 0.0492,
      "step": 16
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.515450119972229,
      "learning_rate": 0.0002,
      "loss": 0.3283,
      "step": 17
    },
    {
      "epoch": 10.285714285714286,
      "grad_norm": 0.6195679903030396,
      "learning_rate": 0.0002,
      "loss": 0.1528,
      "step": 18
    },
    {
      "epoch": 11.0,
      "grad_norm": 0.6195679903030396,
      "learning_rate": 0.0002,
      "loss": 0.1333,
      "step": 19
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": 0.593528151512146,
      "learning_rate": 0.0002,
      "loss": 0.1008,
      "step": 20
    },
    {
      "epoch": 11.428571428571429,
      "eval_loss": 0.1764988899230957,
      "eval_runtime": 1.1181,
      "eval_samples_per_second": 6.26,
      "eval_steps_per_second": 6.26,
      "step": 20
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.593528151512146,
      "learning_rate": 0.0002,
      "loss": 0.1048,
      "step": 21
    },
    {
      "epoch": 12.571428571428571,
      "grad_norm": 0.5447816848754883,
      "learning_rate": 0.0002,
      "loss": 0.1318,
      "step": 22
    },
    {
      "epoch": 13.0,
      "grad_norm": 0.5447816848754883,
      "learning_rate": 0.0002,
      "loss": 0.0098,
      "step": 23
    },
    {
      "epoch": 13.714285714285714,
      "grad_norm": 0.37343642115592957,
      "learning_rate": 0.0002,
      "loss": 0.069,
      "step": 24
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.37343642115592957,
      "learning_rate": 0.0002,
      "loss": 0.0154,
      "step": 25
    },
    {
      "epoch": 14.857142857142858,
      "grad_norm": 0.5212247967720032,
      "learning_rate": 0.0002,
      "loss": 0.0554,
      "step": 26
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.5212247967720032,
      "learning_rate": 0.0002,
      "loss": 0.0011,
      "step": 27
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.39550113677978516,
      "learning_rate": 0.0002,
      "loss": 0.027,
      "step": 28
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.39550113677978516,
      "learning_rate": 0.0002,
      "loss": 0.0154,
      "step": 29
    },
    {
      "epoch": 17.142857142857142,
      "grad_norm": 0.1997860223054886,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 30
    },
    {
      "epoch": 17.142857142857142,
      "eval_loss": 0.009444376453757286,
      "eval_runtime": 1.1094,
      "eval_samples_per_second": 6.309,
      "eval_steps_per_second": 6.309,
      "step": 30
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.1997860223054886,
      "learning_rate": 0.0002,
      "loss": 0.008,
      "step": 31
    },
    {
      "epoch": 18.285714285714285,
      "grad_norm": 0.17780642211437225,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 32
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.17780642211437225,
      "learning_rate": 0.0002,
      "loss": 0.0024,
      "step": 33
    },
    {
      "epoch": 19.428571428571427,
      "grad_norm": 0.09276885539293289,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 34
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.09276885539293289,
      "learning_rate": 0.0002,
      "loss": 0.0008,
      "step": 35
    },
    {
      "epoch": 20.571428571428573,
      "grad_norm": 0.03949622064828873,
      "learning_rate": 0.0002,
      "loss": 0.0004,
      "step": 36
    },
    {
      "epoch": 21.0,
      "grad_norm": 0.03949622064828873,
      "learning_rate": 0.0002,
      "loss": 0.0026,
      "step": 37
    },
    {
      "epoch": 21.714285714285715,
      "grad_norm": 0.48974359035491943,
      "learning_rate": 0.0002,
      "loss": 0.0027,
      "step": 38
    },
    {
      "epoch": 22.0,
      "grad_norm": 0.48974359035491943,
      "learning_rate": 0.0002,
      "loss": 0.0001,
      "step": 39
    },
    {
      "epoch": 22.857142857142858,
      "grad_norm": 0.037879057228565216,
      "learning_rate": 0.0002,
      "loss": 0.0008,
      "step": 40
    },
    {
      "epoch": 22.857142857142858,
      "eval_loss": 0.0010270120110362768,
      "eval_runtime": 1.107,
      "eval_samples_per_second": 6.323,
      "eval_steps_per_second": 6.323,
      "step": 40
    },
    {
      "epoch": 23.0,
      "grad_norm": 0.037879057228565216,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 41
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.07712631672620773,
      "learning_rate": 0.0002,
      "loss": 0.0009,
      "step": 42
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.07712631672620773,
      "learning_rate": 0.0002,
      "loss": 0.0006,
      "step": 43
    },
    {
      "epoch": 25.142857142857142,
      "grad_norm": 0.03069213591516018,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 44
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.03069213591516018,
      "learning_rate": 0.0002,
      "loss": 0.0005,
      "step": 45
    },
    {
      "epoch": 26.285714285714285,
      "grad_norm": 0.04276096820831299,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 46
    },
    {
      "epoch": 27.0,
      "grad_norm": 0.04276096820831299,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 47
    },
    {
      "epoch": 27.428571428571427,
      "grad_norm": 0.018633099272847176,
      "learning_rate": 0.0002,
      "loss": 0.0001,
      "step": 48
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.018633099272847176,
      "learning_rate": 0.0002,
      "loss": 0.0004,
      "step": 49
    },
    {
      "epoch": 28.571428571428573,
      "grad_norm": 0.05017812177538872,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 50
    },
    {
      "epoch": 28.571428571428573,
      "eval_loss": 0.00034485122887417674,
      "eval_runtime": 1.12,
      "eval_samples_per_second": 6.25,
      "eval_steps_per_second": 6.25,
      "step": 50
    },
    {
      "epoch": 29.0,
      "grad_norm": 0.05017812177538872,
      "learning_rate": 0.0002,
      "loss": 0.0001,
      "step": 51
    },
    {
      "epoch": 29.714285714285715,
      "grad_norm": 0.010933789424598217,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 52
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.010933789424598217,
      "learning_rate": 0.0002,
      "loss": 0.0001,
      "step": 53
    },
    {
      "epoch": 30.857142857142858,
      "grad_norm": 0.00900160800665617,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 54
    },
    {
      "epoch": 31.0,
      "grad_norm": 0.00900160800665617,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 55
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.018903745338320732,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 56
    },
    {
      "epoch": 33.0,
      "grad_norm": 0.018903745338320732,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 57
    },
    {
      "epoch": 33.142857142857146,
      "grad_norm": 0.01036656741052866,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 58
    },
    {
      "epoch": 34.0,
      "grad_norm": 0.01036656741052866,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 59
    },
    {
      "epoch": 34.285714285714285,
      "grad_norm": 0.007697099819779396,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 60
    },
    {
      "epoch": 34.285714285714285,
      "eval_loss": 0.00019181256357114762,
      "eval_runtime": 1.1116,
      "eval_samples_per_second": 6.297,
      "eval_steps_per_second": 6.297,
      "step": 60
    },
    {
      "epoch": 34.285714285714285,
      "step": 60,
      "total_flos": 4016214881861632.0,
      "train_loss": 0.1552126432957569,
      "train_runtime": 117.5216,
      "train_samples_per_second": 4.084,
      "train_steps_per_second": 0.511
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 60,
  "save_steps": 1000,
  "total_flos": 4016214881861632.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}