{
  "best_metric": 0.4628298878669739,
  "best_model_checkpoint": "mikhail_panzo/zlm_b64_le5_s12000/checkpoint-1500",
  "epoch": 1.256544502617801,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.041884816753926704,
      "grad_norm": 12.821951866149902,
      "learning_rate": 2.4000000000000003e-07,
      "loss": 1.1152,
      "step": 50
    },
    {
      "epoch": 0.08376963350785341,
      "grad_norm": 6.186221122741699,
      "learning_rate": 4.900000000000001e-07,
      "loss": 1.0584,
      "step": 100
    },
    {
      "epoch": 0.1256544502617801,
      "grad_norm": 13.08674430847168,
      "learning_rate": 7.4e-07,
      "loss": 0.9996,
      "step": 150
    },
    {
      "epoch": 0.16753926701570682,
      "grad_norm": 4.10632848739624,
      "learning_rate": 9.9e-07,
      "loss": 0.9355,
      "step": 200
    },
    {
      "epoch": 0.2094240837696335,
      "grad_norm": 3.2428064346313477,
      "learning_rate": 1.2400000000000002e-06,
      "loss": 0.852,
      "step": 250
    },
    {
      "epoch": 0.2513089005235602,
      "grad_norm": 2.0861401557922363,
      "learning_rate": 1.4900000000000001e-06,
      "loss": 0.8241,
      "step": 300
    },
    {
      "epoch": 0.2931937172774869,
      "grad_norm": 8.602317810058594,
      "learning_rate": 1.74e-06,
      "loss": 0.7629,
      "step": 350
    },
    {
      "epoch": 0.33507853403141363,
      "grad_norm": 2.7594990730285645,
      "learning_rate": 1.9900000000000004e-06,
      "loss": 0.7418,
      "step": 400
    },
    {
      "epoch": 0.3769633507853403,
      "grad_norm": 2.7047348022460938,
      "learning_rate": 2.24e-06,
      "loss": 0.7166,
      "step": 450
    },
    {
      "epoch": 0.418848167539267,
      "grad_norm": 2.0666158199310303,
      "learning_rate": 2.4900000000000003e-06,
      "loss": 0.6876,
      "step": 500
    },
    {
      "epoch": 0.418848167539267,
      "eval_loss": 0.5937426090240479,
      "eval_runtime": 263.0955,
      "eval_samples_per_second": 32.266,
      "eval_steps_per_second": 4.037,
      "step": 500
    },
    {
      "epoch": 0.4607329842931937,
      "grad_norm": 2.955193519592285,
      "learning_rate": 2.7400000000000004e-06,
      "loss": 0.673,
      "step": 550
    },
    {
      "epoch": 0.5026178010471204,
      "grad_norm": 4.856322765350342,
      "learning_rate": 2.99e-06,
      "loss": 0.6539,
      "step": 600
    },
    {
      "epoch": 0.5445026178010471,
      "grad_norm": 7.631103992462158,
      "learning_rate": 3.2400000000000003e-06,
      "loss": 0.6588,
      "step": 650
    },
    {
      "epoch": 0.5863874345549738,
      "grad_norm": 2.715771436691284,
      "learning_rate": 3.49e-06,
      "loss": 0.624,
      "step": 700
    },
    {
      "epoch": 0.6282722513089005,
      "grad_norm": 2.7684273719787598,
      "learning_rate": 3.74e-06,
      "loss": 0.6116,
      "step": 750
    },
    {
      "epoch": 0.6701570680628273,
      "grad_norm": 2.219599723815918,
      "learning_rate": 3.990000000000001e-06,
      "loss": 0.6097,
      "step": 800
    },
    {
      "epoch": 0.7120418848167539,
      "grad_norm": 3.749188184738159,
      "learning_rate": 4.24e-06,
      "loss": 0.614,
      "step": 850
    },
    {
      "epoch": 0.7539267015706806,
      "grad_norm": 2.120619773864746,
      "learning_rate": 4.49e-06,
      "loss": 0.5814,
      "step": 900
    },
    {
      "epoch": 0.7958115183246073,
      "grad_norm": 2.7725327014923096,
      "learning_rate": 4.74e-06,
      "loss": 0.5895,
      "step": 950
    },
    {
      "epoch": 0.837696335078534,
      "grad_norm": 9.851961135864258,
      "learning_rate": 4.9900000000000005e-06,
      "loss": 0.5623,
      "step": 1000
    },
    {
      "epoch": 0.837696335078534,
      "eval_loss": 0.4933945834636688,
      "eval_runtime": 258.739,
      "eval_samples_per_second": 32.809,
      "eval_steps_per_second": 4.105,
      "step": 1000
    },
    {
      "epoch": 0.8795811518324608,
      "grad_norm": 5.074089527130127,
      "learning_rate": 5.240000000000001e-06,
      "loss": 0.5748,
      "step": 1050
    },
    {
      "epoch": 0.9214659685863874,
      "grad_norm": 5.884639739990234,
      "learning_rate": 5.485e-06,
      "loss": 0.5395,
      "step": 1100
    },
    {
      "epoch": 0.9633507853403142,
      "grad_norm": 3.1588447093963623,
      "learning_rate": 5.735e-06,
      "loss": 0.5478,
      "step": 1150
    },
    {
      "epoch": 1.0052356020942408,
      "grad_norm": 2.4452970027923584,
      "learning_rate": 5.985000000000001e-06,
      "loss": 0.538,
      "step": 1200
    },
    {
      "epoch": 1.0471204188481675,
      "grad_norm": 8.290769577026367,
      "learning_rate": 6.235000000000001e-06,
      "loss": 0.5411,
      "step": 1250
    },
    {
      "epoch": 1.0890052356020943,
      "grad_norm": 4.080046653747559,
      "learning_rate": 6.485000000000001e-06,
      "loss": 0.5387,
      "step": 1300
    },
    {
      "epoch": 1.130890052356021,
      "grad_norm": 2.7520201206207275,
      "learning_rate": 6.735000000000001e-06,
      "loss": 0.5426,
      "step": 1350
    },
    {
      "epoch": 1.1727748691099475,
      "grad_norm": 3.8192107677459717,
      "learning_rate": 6.985000000000001e-06,
      "loss": 0.5315,
      "step": 1400
    },
    {
      "epoch": 1.2146596858638743,
      "grad_norm": 9.535676956176758,
      "learning_rate": 7.235000000000001e-06,
      "loss": 0.5159,
      "step": 1450
    },
    {
      "epoch": 1.256544502617801,
      "grad_norm": 2.5731022357940674,
      "learning_rate": 7.485000000000001e-06,
      "loss": 0.5165,
      "step": 1500
    },
    {
      "epoch": 1.256544502617801,
      "eval_loss": 0.4628298878669739,
      "eval_runtime": 257.0265,
      "eval_samples_per_second": 33.028,
      "eval_steps_per_second": 4.132,
      "step": 1500
    }
  ],
  "logging_steps": 50,
  "max_steps": 12000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 11,
  "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": 1.3505481451642752e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}