{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.62962962962963,
  "eval_steps": 500,
  "global_step": 130,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07407407407407407,
      "grad_norm": 5.0625,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 3.0102,
      "step": 1
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 2.390625,
      "learning_rate": 7.692307692307693e-05,
      "loss": 3.0215,
      "step": 5
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 11.5,
      "learning_rate": 0.00015384615384615385,
      "loss": 2.7544,
      "step": 10
    },
    {
      "epoch": 0.9629629629629629,
      "eval_loss": 2.8721797466278076,
      "eval_runtime": 0.652,
      "eval_samples_per_second": 15.338,
      "eval_steps_per_second": 1.534,
      "step": 13
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019985583705641418,
      "loss": 2.379,
      "step": 15
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 2.609375,
      "learning_rate": 0.00019823877374156647,
      "loss": 2.0489,
      "step": 20
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019485364419471454,
      "loss": 1.7723,
      "step": 25
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.60640549659729,
      "eval_runtime": 0.5469,
      "eval_samples_per_second": 18.283,
      "eval_steps_per_second": 1.828,
      "step": 27
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0001897613727639014,
      "loss": 1.6022,
      "step": 30
    },
    {
      "epoch": 2.5925925925925926,
      "grad_norm": 0.65234375,
      "learning_rate": 0.00018305360832480117,
      "loss": 1.4855,
      "step": 35
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 0.58203125,
      "learning_rate": 0.00017485107481711012,
      "loss": 1.4023,
      "step": 40
    },
    {
      "epoch": 2.962962962962963,
      "eval_loss": 2.5709500312805176,
      "eval_runtime": 0.7217,
      "eval_samples_per_second": 13.857,
      "eval_steps_per_second": 1.386,
      "step": 40
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0001653013984983585,
      "loss": 1.3253,
      "step": 45
    },
    {
      "epoch": 3.7037037037037037,
      "grad_norm": 0.50390625,
      "learning_rate": 0.00015457645101945046,
      "loss": 1.2778,
      "step": 50
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.5349316596984863,
      "eval_runtime": 0.547,
      "eval_samples_per_second": 18.283,
      "eval_steps_per_second": 1.828,
      "step": 54
    },
    {
      "epoch": 4.074074074074074,
      "grad_norm": 0.376953125,
      "learning_rate": 0.00014286925614030542,
      "loss": 1.2498,
      "step": 55
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0001303905157574247,
      "loss": 1.2221,
      "step": 60
    },
    {
      "epoch": 4.814814814814815,
      "grad_norm": 0.296875,
      "learning_rate": 0.00011736481776669306,
      "loss": 1.1848,
      "step": 65
    },
    {
      "epoch": 4.962962962962963,
      "eval_loss": 2.5175788402557373,
      "eval_runtime": 0.6693,
      "eval_samples_per_second": 14.942,
      "eval_steps_per_second": 1.494,
      "step": 67
    },
    {
      "epoch": 5.185185185185185,
      "grad_norm": 0.291015625,
      "learning_rate": 0.00010402659401094152,
      "loss": 1.1814,
      "step": 70
    },
    {
      "epoch": 5.555555555555555,
      "grad_norm": 0.33203125,
      "learning_rate": 9.061590105968208e-05,
      "loss": 1.1574,
      "step": 75
    },
    {
      "epoch": 5.925925925925926,
      "grad_norm": 0.349609375,
      "learning_rate": 7.73740997570278e-05,
      "loss": 1.1522,
      "step": 80
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.5044538974761963,
      "eval_runtime": 0.5444,
      "eval_samples_per_second": 18.369,
      "eval_steps_per_second": 1.837,
      "step": 81
    },
    {
      "epoch": 6.296296296296296,
      "grad_norm": 0.2412109375,
      "learning_rate": 6.453951129574644e-05,
      "loss": 1.1367,
      "step": 85
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.28515625,
      "learning_rate": 5.234312799786921e-05,
      "loss": 1.1305,
      "step": 90
    },
    {
      "epoch": 6.962962962962963,
      "eval_loss": 2.506514310836792,
      "eval_runtime": 0.685,
      "eval_samples_per_second": 14.598,
      "eval_steps_per_second": 1.46,
      "step": 94
    },
    {
      "epoch": 7.037037037037037,
      "grad_norm": 0.2265625,
      "learning_rate": 4.100445599768774e-05,
      "loss": 1.1188,
      "step": 95
    },
    {
      "epoch": 7.407407407407407,
      "grad_norm": 0.2099609375,
      "learning_rate": 3.072756464904006e-05,
      "loss": 1.1222,
      "step": 100
    },
    {
      "epoch": 7.777777777777778,
      "grad_norm": 0.248046875,
      "learning_rate": 2.1697413758237784e-05,
      "loss": 1.1075,
      "step": 105
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.5136494636535645,
      "eval_runtime": 0.5462,
      "eval_samples_per_second": 18.307,
      "eval_steps_per_second": 1.831,
      "step": 108
    },
    {
      "epoch": 8.148148148148149,
      "grad_norm": 0.2470703125,
      "learning_rate": 1.4076524743778319e-05,
      "loss": 1.1126,
      "step": 110
    },
    {
      "epoch": 8.518518518518519,
      "grad_norm": 0.236328125,
      "learning_rate": 8.002055634117578e-06,
      "loss": 1.1118,
      "step": 115
    },
    {
      "epoch": 8.88888888888889,
      "grad_norm": 0.2275390625,
      "learning_rate": 3.5833325466437694e-06,
      "loss": 1.1049,
      "step": 120
    },
    {
      "epoch": 8.962962962962964,
      "eval_loss": 2.512882709503174,
      "eval_runtime": 0.6584,
      "eval_samples_per_second": 15.188,
      "eval_steps_per_second": 1.519,
      "step": 121
    },
    {
      "epoch": 9.25925925925926,
      "grad_norm": 0.2216796875,
      "learning_rate": 8.998820754091531e-07,
      "loss": 1.1116,
      "step": 125
    },
    {
      "epoch": 9.62962962962963,
      "grad_norm": 0.20703125,
      "learning_rate": 0.0,
      "loss": 1.1048,
      "step": 130
    },
    {
      "epoch": 9.62962962962963,
      "eval_loss": 2.513336658477783,
      "eval_runtime": 0.5375,
      "eval_samples_per_second": 18.606,
      "eval_steps_per_second": 1.861,
      "step": 130
    },
    {
      "epoch": 9.62962962962963,
      "step": 130,
      "total_flos": 1.018113810235392e+17,
      "train_loss": 1.4375356710874116,
      "train_runtime": 456.268,
      "train_samples_per_second": 18.301,
      "train_steps_per_second": 0.285
    }
  ],
  "logging_steps": 5,
  "max_steps": 130,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "total_flos": 1.018113810235392e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}