{
  "best_metric": 2.4261486530303955,
  "best_model_checkpoint": "./output/training_results/C016_Meta-Llama-3-8B_pretrain_20240721_092214/checkpoint-11088",
  "epoch": 4.0,
  "eval_steps": 1232,
  "global_step": 12316,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003247807729782397,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.6721,
      "step": 1
    },
    {
      "epoch": 0.20006495615459566,
      "grad_norm": 1.9328745806142353,
      "learning_rate": 1.9805194805194805e-06,
      "loss": 2.592,
      "step": 616
    },
    {
      "epoch": 0.4001299123091913,
      "grad_norm": 2.0226924668887496,
      "learning_rate": 2.245175689219919e-06,
      "loss": 2.5057,
      "step": 1232
    },
    {
      "epoch": 0.4001299123091913,
      "eval_loss": 2.4827427864074707,
      "eval_runtime": 252.9711,
      "eval_samples_per_second": 86.543,
      "eval_steps_per_second": 0.68,
      "step": 1232
    },
    {
      "epoch": 0.6001948684637869,
      "grad_norm": 1.8512197610266496,
      "learning_rate": 1.2232016471327423e-06,
      "loss": 2.4683,
      "step": 1848
    },
    {
      "epoch": 0.8002598246183826,
      "grad_norm": 1.989781044650789,
      "learning_rate": 6.53049308175953e-07,
      "loss": 2.444,
      "step": 2464
    },
    {
      "epoch": 0.8002598246183826,
      "eval_loss": 2.440027952194214,
      "eval_runtime": 251.3004,
      "eval_samples_per_second": 87.119,
      "eval_steps_per_second": 0.684,
      "step": 2464
    },
    {
      "epoch": 1.0003247807729783,
      "grad_norm": 1.8956736763179278,
      "learning_rate": 3.466185770829244e-07,
      "loss": 2.4358,
      "step": 3080
    },
    {
      "epoch": 1.200389736927574,
      "grad_norm": 1.9111043304334017,
      "learning_rate": 1.8910445197889315e-07,
      "loss": 2.3648,
      "step": 3696
    },
    {
      "epoch": 1.200389736927574,
      "eval_loss": 2.431915760040283,
      "eval_runtime": 251.1577,
      "eval_samples_per_second": 87.168,
      "eval_steps_per_second": 0.685,
      "step": 3696
    },
    {
      "epoch": 1.4004546930821695,
      "grad_norm": 1.884967762986231,
      "learning_rate": 1.1168237259086467e-07,
      "loss": 2.3663,
      "step": 4312
    },
    {
      "epoch": 1.600519649236765,
      "grad_norm": 1.9873340014256546,
      "learning_rate": 7.563133304849047e-08,
      "loss": 2.372,
      "step": 4928
    },
    {
      "epoch": 1.600519649236765,
      "eval_loss": 2.4293837547302246,
      "eval_runtime": 251.386,
      "eval_samples_per_second": 87.089,
      "eval_steps_per_second": 0.684,
      "step": 4928
    },
    {
      "epoch": 1.8005846053913608,
      "grad_norm": 1.9124289441200373,
      "learning_rate": 5.98689431836726e-08,
      "loss": 2.3684,
      "step": 5544
    },
    {
      "epoch": 2.0006495615459565,
      "grad_norm": 1.9479679838446426,
      "learning_rate": 5.346405476547749e-08,
      "loss": 2.3667,
      "step": 6160
    },
    {
      "epoch": 2.0006495615459565,
      "eval_loss": 2.4281327724456787,
      "eval_runtime": 251.4248,
      "eval_samples_per_second": 87.076,
      "eval_steps_per_second": 0.684,
      "step": 6160
    },
    {
      "epoch": 2.2007145177005523,
      "grad_norm": 1.9022764488275758,
      "learning_rate": 5.109115615383696e-08,
      "loss": 2.3568,
      "step": 6776
    },
    {
      "epoch": 2.400779473855148,
      "grad_norm": 1.885664698080457,
      "learning_rate": 5.0300090028337e-08,
      "loss": 2.3573,
      "step": 7392
    },
    {
      "epoch": 2.400779473855148,
      "eval_loss": 2.4281272888183594,
      "eval_runtime": 250.732,
      "eval_samples_per_second": 87.316,
      "eval_steps_per_second": 0.686,
      "step": 7392
    },
    {
      "epoch": 2.6008444300097433,
      "grad_norm": 1.9790266878041507,
      "learning_rate": 5.006932020966859e-08,
      "loss": 2.3533,
      "step": 8008
    },
    {
      "epoch": 2.800909386164339,
      "grad_norm": 1.9987792342132904,
      "learning_rate": 5.0012816199435985e-08,
      "loss": 2.3603,
      "step": 8624
    },
    {
      "epoch": 2.800909386164339,
      "eval_loss": 2.4273290634155273,
      "eval_runtime": 251.237,
      "eval_samples_per_second": 87.141,
      "eval_steps_per_second": 0.685,
      "step": 8624
    },
    {
      "epoch": 3.000974342318935,
      "grad_norm": 1.892342522891005,
      "learning_rate": 5.0001737227175665e-08,
      "loss": 2.3575,
      "step": 9240
    },
    {
      "epoch": 3.2010392984735305,
      "grad_norm": 1.945849916471603,
      "learning_rate": 5.0000151498505686e-08,
      "loss": 2.3522,
      "step": 9856
    },
    {
      "epoch": 3.2010392984735305,
      "eval_loss": 2.4268627166748047,
      "eval_runtime": 251.3708,
      "eval_samples_per_second": 87.094,
      "eval_steps_per_second": 0.684,
      "step": 9856
    },
    {
      "epoch": 3.401104254628126,
      "grad_norm": 1.9950133710544578,
      "learning_rate": 5.0000006522774745e-08,
      "loss": 2.3532,
      "step": 10472
    },
    {
      "epoch": 3.6011692107827216,
      "grad_norm": 1.9873010505050475,
      "learning_rate": 5.000000007975414e-08,
      "loss": 2.353,
      "step": 11088
    },
    {
      "epoch": 3.6011692107827216,
      "eval_loss": 2.4261486530303955,
      "eval_runtime": 252.6429,
      "eval_samples_per_second": 86.656,
      "eval_steps_per_second": 0.681,
      "step": 11088
    },
    {
      "epoch": 3.8012341669373173,
      "grad_norm": 1.998167327050962,
      "learning_rate": 5.0000000000044414e-08,
      "loss": 2.3517,
      "step": 11704
    },
    {
      "epoch": 4.0,
      "step": 12316,
      "total_flos": 1287266123120640.0,
      "train_loss": 2.391189053524001,
      "train_runtime": 40307.7885,
      "train_samples_per_second": 19.553,
      "train_steps_per_second": 0.306
    }
  ],
  "logging_steps": 616,
  "max_steps": 12316,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 1232,
  "total_flos": 1287266123120640.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}