{
  "best_metric": 0.3777858018875122,
  "best_model_checkpoint": "mikhail-panzo/zlm_b128_le4_s4000/checkpoint-1500",
  "epoch": 2.513089005235602,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08376963350785341,
      "grad_norm": 2.9717624187469482,
      "learning_rate": 2.4500000000000003e-06,
      "loss": 1.0424,
      "step": 50
    },
    {
      "epoch": 0.16753926701570682,
      "grad_norm": 2.9720630645751953,
      "learning_rate": 4.950000000000001e-06,
      "loss": 0.8474,
      "step": 100
    },
    {
      "epoch": 0.2513089005235602,
      "grad_norm": 2.445929765701294,
      "learning_rate": 7.45e-06,
      "loss": 0.7336,
      "step": 150
    },
    {
      "epoch": 0.33507853403141363,
      "grad_norm": 5.502955913543701,
      "learning_rate": 9.950000000000001e-06,
      "loss": 0.6492,
      "step": 200
    },
    {
      "epoch": 0.418848167539267,
      "grad_norm": 2.3356130123138428,
      "learning_rate": 1.2450000000000001e-05,
      "loss": 0.6133,
      "step": 250
    },
    {
      "epoch": 0.5026178010471204,
      "grad_norm": 1.937270164489746,
      "learning_rate": 1.4950000000000001e-05,
      "loss": 0.5889,
      "step": 300
    },
    {
      "epoch": 0.5863874345549738,
      "grad_norm": 2.392244338989258,
      "learning_rate": 1.745e-05,
      "loss": 0.5694,
      "step": 350
    },
    {
      "epoch": 0.6701570680628273,
      "grad_norm": 7.3209919929504395,
      "learning_rate": 1.995e-05,
      "loss": 0.5477,
      "step": 400
    },
    {
      "epoch": 0.7539267015706806,
      "grad_norm": 3.415917158126831,
      "learning_rate": 2.245e-05,
      "loss": 0.5329,
      "step": 450
    },
    {
      "epoch": 0.837696335078534,
      "grad_norm": 3.0256705284118652,
      "learning_rate": 2.495e-05,
      "loss": 0.5173,
      "step": 500
    },
    {
      "epoch": 0.837696335078534,
      "eval_loss": 0.4566049873828888,
      "eval_runtime": 261.3511,
      "eval_samples_per_second": 32.481,
      "eval_steps_per_second": 4.063,
      "step": 500
    },
    {
      "epoch": 0.9214659685863874,
      "grad_norm": 1.9436837434768677,
      "learning_rate": 2.7450000000000003e-05,
      "loss": 0.5079,
      "step": 550
    },
    {
      "epoch": 1.0052356020942408,
      "grad_norm": 1.819956660270691,
      "learning_rate": 2.995e-05,
      "loss": 0.4969,
      "step": 600
    },
    {
      "epoch": 1.0890052356020943,
      "grad_norm": 5.457251071929932,
      "learning_rate": 3.245e-05,
      "loss": 0.4977,
      "step": 650
    },
    {
      "epoch": 1.1727748691099475,
      "grad_norm": 3.183980703353882,
      "learning_rate": 3.495e-05,
      "loss": 0.4923,
      "step": 700
    },
    {
      "epoch": 1.256544502617801,
      "grad_norm": 7.1660051345825195,
      "learning_rate": 3.745e-05,
      "loss": 0.4802,
      "step": 750
    },
    {
      "epoch": 1.3403141361256545,
      "grad_norm": 5.499026775360107,
      "learning_rate": 3.995e-05,
      "loss": 0.4754,
      "step": 800
    },
    {
      "epoch": 1.4240837696335078,
      "grad_norm": 2.8053908348083496,
      "learning_rate": 4.245e-05,
      "loss": 0.4669,
      "step": 850
    },
    {
      "epoch": 1.5078534031413613,
      "grad_norm": 3.017005443572998,
      "learning_rate": 4.495e-05,
      "loss": 0.4604,
      "step": 900
    },
    {
      "epoch": 1.5916230366492146,
      "grad_norm": 2.7971177101135254,
      "learning_rate": 4.745e-05,
      "loss": 0.4565,
      "step": 950
    },
    {
      "epoch": 1.675392670157068,
      "grad_norm": 3.1588356494903564,
      "learning_rate": 4.995e-05,
      "loss": 0.455,
      "step": 1000
    },
    {
      "epoch": 1.675392670157068,
      "eval_loss": 0.40312233567237854,
      "eval_runtime": 256.4334,
      "eval_samples_per_second": 33.104,
      "eval_steps_per_second": 4.141,
      "step": 1000
    },
    {
      "epoch": 1.7591623036649215,
      "grad_norm": 2.2053232192993164,
      "learning_rate": 5.245e-05,
      "loss": 0.4543,
      "step": 1050
    },
    {
      "epoch": 1.8429319371727748,
      "grad_norm": 2.0562164783477783,
      "learning_rate": 5.495e-05,
      "loss": 0.4456,
      "step": 1100
    },
    {
      "epoch": 1.9267015706806283,
      "grad_norm": 2.730119466781616,
      "learning_rate": 5.745e-05,
      "loss": 0.4355,
      "step": 1150
    },
    {
      "epoch": 2.0104712041884816,
      "grad_norm": 1.7484283447265625,
      "learning_rate": 5.995000000000001e-05,
      "loss": 0.4299,
      "step": 1200
    },
    {
      "epoch": 2.094240837696335,
      "grad_norm": 1.1786061525344849,
      "learning_rate": 6.245000000000001e-05,
      "loss": 0.4305,
      "step": 1250
    },
    {
      "epoch": 2.1780104712041886,
      "grad_norm": 1.98978590965271,
      "learning_rate": 6.494999999999999e-05,
      "loss": 0.4295,
      "step": 1300
    },
    {
      "epoch": 2.261780104712042,
      "grad_norm": 2.818659782409668,
      "learning_rate": 6.745e-05,
      "loss": 0.4235,
      "step": 1350
    },
    {
      "epoch": 2.345549738219895,
      "grad_norm": 2.3864262104034424,
      "learning_rate": 6.995e-05,
      "loss": 0.4271,
      "step": 1400
    },
    {
      "epoch": 2.4293193717277486,
      "grad_norm": 1.3647903203964233,
      "learning_rate": 7.245000000000001e-05,
      "loss": 0.4208,
      "step": 1450
    },
    {
      "epoch": 2.513089005235602,
      "grad_norm": 2.2144172191619873,
      "learning_rate": 7.495e-05,
      "loss": 0.4175,
      "step": 1500
    },
    {
      "epoch": 2.513089005235602,
      "eval_loss": 0.3777858018875122,
      "eval_runtime": 260.5025,
      "eval_samples_per_second": 32.587,
      "eval_steps_per_second": 4.077,
      "step": 1500
    }
  ],
  "logging_steps": 50,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 2.686672014814656e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}