{
  "best_metric": 0.4093586802482605,
  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s12000/checkpoint-1000",
  "epoch": 1.675392670157068,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08376963350785341,
      "grad_norm": 2.469545841217041,
      "learning_rate": 2.4500000000000003e-06,
      "loss": 1.0402,
      "step": 50
    },
    {
      "epoch": 0.16753926701570682,
      "grad_norm": 3.3642780780792236,
      "learning_rate": 4.950000000000001e-06,
      "loss": 0.8472,
      "step": 100
    },
    {
      "epoch": 0.2513089005235602,
      "grad_norm": 1.8375087976455688,
      "learning_rate": 7.45e-06,
      "loss": 0.7331,
      "step": 150
    },
    {
      "epoch": 0.33507853403141363,
      "grad_norm": 3.700824737548828,
      "learning_rate": 9.950000000000001e-06,
      "loss": 0.6472,
      "step": 200
    },
    {
      "epoch": 0.418848167539267,
      "grad_norm": 1.7632888555526733,
      "learning_rate": 1.2450000000000001e-05,
      "loss": 0.6112,
      "step": 250
    },
    {
      "epoch": 0.5026178010471204,
      "grad_norm": 3.1861908435821533,
      "learning_rate": 1.4950000000000001e-05,
      "loss": 0.5865,
      "step": 300
    },
    {
      "epoch": 0.5863874345549738,
      "grad_norm": 2.2845046520233154,
      "learning_rate": 1.745e-05,
      "loss": 0.5682,
      "step": 350
    },
    {
      "epoch": 0.6701570680628273,
      "grad_norm": 2.079210042953491,
      "learning_rate": 1.995e-05,
      "loss": 0.5465,
      "step": 400
    },
    {
      "epoch": 0.7539267015706806,
      "grad_norm": 3.669891119003296,
      "learning_rate": 2.245e-05,
      "loss": 0.5302,
      "step": 450
    },
    {
      "epoch": 0.837696335078534,
      "grad_norm": 2.4679417610168457,
      "learning_rate": 2.495e-05,
      "loss": 0.5137,
      "step": 500
    },
    {
      "epoch": 0.837696335078534,
      "eval_loss": 0.4513999819755554,
      "eval_runtime": 277.1074,
      "eval_samples_per_second": 30.634,
      "eval_steps_per_second": 3.832,
      "step": 500
    },
    {
      "epoch": 0.9214659685863874,
      "grad_norm": 2.703871726989746,
      "learning_rate": 2.7450000000000003e-05,
      "loss": 0.5071,
      "step": 550
    },
    {
      "epoch": 1.0052356020942408,
      "grad_norm": 3.8184635639190674,
      "learning_rate": 2.995e-05,
      "loss": 0.4971,
      "step": 600
    },
    {
      "epoch": 1.0890052356020943,
      "grad_norm": 2.2857866287231445,
      "learning_rate": 3.245e-05,
      "loss": 0.4955,
      "step": 650
    },
    {
      "epoch": 1.1727748691099475,
      "grad_norm": 3.5974085330963135,
      "learning_rate": 3.495e-05,
      "loss": 0.4935,
      "step": 700
    },
    {
      "epoch": 1.256544502617801,
      "grad_norm": 1.720818281173706,
      "learning_rate": 3.745e-05,
      "loss": 0.4763,
      "step": 750
    },
    {
      "epoch": 1.3403141361256545,
      "grad_norm": 3.080139636993408,
      "learning_rate": 3.995e-05,
      "loss": 0.4735,
      "step": 800
    },
    {
      "epoch": 1.4240837696335078,
      "grad_norm": 4.877579212188721,
      "learning_rate": 4.245e-05,
      "loss": 0.4654,
      "step": 850
    },
    {
      "epoch": 1.5078534031413613,
      "grad_norm": 3.383965253829956,
      "learning_rate": 4.495e-05,
      "loss": 0.4628,
      "step": 900
    },
    {
      "epoch": 1.5916230366492146,
      "grad_norm": 3.3636982440948486,
      "learning_rate": 4.745e-05,
      "loss": 0.4541,
      "step": 950
    },
    {
      "epoch": 1.675392670157068,
      "grad_norm": 1.666568398475647,
      "learning_rate": 4.995e-05,
      "loss": 0.4565,
      "step": 1000
    },
    {
      "epoch": 1.675392670157068,
      "eval_loss": 0.4093586802482605,
      "eval_runtime": 269.4005,
      "eval_samples_per_second": 31.511,
      "eval_steps_per_second": 3.942,
      "step": 1000
    }
  ],
  "logging_steps": 50,
  "max_steps": 12000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 21,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7957703358412928e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}