{
  "best_metric": 0.4592318534851074,
  "best_model_checkpoint": "mikhail-panzo/zlm_b64_le5_s8000/checkpoint-1500",
  "epoch": 1.2562814070351758,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04187604690117253,
      "grad_norm": 5.415271759033203,
      "learning_rate": 2.4000000000000003e-07,
      "loss": 1.1819,
      "step": 50
    },
    {
      "epoch": 0.08375209380234507,
      "grad_norm": 16.256921768188477,
      "learning_rate": 4.900000000000001e-07,
      "loss": 1.0035,
      "step": 100
    },
    {
      "epoch": 0.12562814070351758,
      "grad_norm": 3.2439260482788086,
      "learning_rate": 7.350000000000001e-07,
      "loss": 0.9742,
      "step": 150
    },
    {
      "epoch": 0.16750418760469013,
      "grad_norm": 4.24526834487915,
      "learning_rate": 9.85e-07,
      "loss": 0.9361,
      "step": 200
    },
    {
      "epoch": 0.20938023450586266,
      "grad_norm": 3.625537872314453,
      "learning_rate": 1.235e-06,
      "loss": 0.8733,
      "step": 250
    },
    {
      "epoch": 0.25125628140703515,
      "grad_norm": 4.1455464363098145,
      "learning_rate": 1.485e-06,
      "loss": 0.8471,
      "step": 300
    },
    {
      "epoch": 0.2931323283082077,
      "grad_norm": 3.164158582687378,
      "learning_rate": 1.7350000000000001e-06,
      "loss": 0.838,
      "step": 350
    },
    {
      "epoch": 0.33500837520938026,
      "grad_norm": 4.24931001663208,
      "learning_rate": 1.985e-06,
      "loss": 0.7467,
      "step": 400
    },
    {
      "epoch": 0.3768844221105528,
      "grad_norm": 2.332688331604004,
      "learning_rate": 2.235e-06,
      "loss": 0.7389,
      "step": 450
    },
    {
      "epoch": 0.4187604690117253,
      "grad_norm": 2.370961904525757,
      "learning_rate": 2.4850000000000003e-06,
      "loss": 0.7074,
      "step": 500
    },
    {
      "epoch": 0.4187604690117253,
      "eval_loss": 0.6029486060142517,
      "eval_runtime": 220.2253,
      "eval_samples_per_second": 38.547,
      "eval_steps_per_second": 4.822,
      "step": 500
    },
    {
      "epoch": 0.46063651591289784,
      "grad_norm": 14.466177940368652,
      "learning_rate": 2.7350000000000005e-06,
      "loss": 0.6738,
      "step": 550
    },
    {
      "epoch": 0.5025125628140703,
      "grad_norm": 3.282449722290039,
      "learning_rate": 2.9850000000000002e-06,
      "loss": 0.6517,
      "step": 600
    },
    {
      "epoch": 0.5443886097152428,
      "grad_norm": 2.621645212173462,
      "learning_rate": 3.2350000000000004e-06,
      "loss": 0.6479,
      "step": 650
    },
    {
      "epoch": 0.5862646566164154,
      "grad_norm": 2.4493019580841064,
      "learning_rate": 3.485e-06,
      "loss": 0.6321,
      "step": 700
    },
    {
      "epoch": 0.628140703517588,
      "grad_norm": 1.919048547744751,
      "learning_rate": 3.7350000000000002e-06,
      "loss": 0.6407,
      "step": 750
    },
    {
      "epoch": 0.6700167504187605,
      "grad_norm": 2.8260602951049805,
      "learning_rate": 3.985000000000001e-06,
      "loss": 0.6279,
      "step": 800
    },
    {
      "epoch": 0.711892797319933,
      "grad_norm": 3.333841323852539,
      "learning_rate": 4.235e-06,
      "loss": 0.6143,
      "step": 850
    },
    {
      "epoch": 0.7537688442211056,
      "grad_norm": 1.7883094549179077,
      "learning_rate": 4.485e-06,
      "loss": 0.5949,
      "step": 900
    },
    {
      "epoch": 0.7956448911222781,
      "grad_norm": 2.48634934425354,
      "learning_rate": 4.735e-06,
      "loss": 0.5875,
      "step": 950
    },
    {
      "epoch": 0.8375209380234506,
      "grad_norm": 5.4277873039245605,
      "learning_rate": 4.9850000000000006e-06,
      "loss": 0.5916,
      "step": 1000
    },
    {
      "epoch": 0.8375209380234506,
      "eval_loss": 0.49684029817581177,
      "eval_runtime": 219.2621,
      "eval_samples_per_second": 38.716,
      "eval_steps_per_second": 4.844,
      "step": 1000
    },
    {
      "epoch": 0.8793969849246231,
      "grad_norm": 2.675995349884033,
      "learning_rate": 5.235e-06,
      "loss": 0.5676,
      "step": 1050
    },
    {
      "epoch": 0.9212730318257957,
      "grad_norm": 2.507474422454834,
      "learning_rate": 5.485e-06,
      "loss": 0.5535,
      "step": 1100
    },
    {
      "epoch": 0.9631490787269682,
      "grad_norm": 5.360454082489014,
      "learning_rate": 5.735e-06,
      "loss": 0.569,
      "step": 1150
    },
    {
      "epoch": 1.0050251256281406,
      "grad_norm": 2.9445533752441406,
      "learning_rate": 5.985000000000001e-06,
      "loss": 0.5385,
      "step": 1200
    },
    {
      "epoch": 1.0469011725293131,
      "grad_norm": 2.3764073848724365,
      "learning_rate": 6.235000000000001e-06,
      "loss": 0.5647,
      "step": 1250
    },
    {
      "epoch": 1.0887772194304857,
      "grad_norm": 4.364608287811279,
      "learning_rate": 6.485000000000001e-06,
      "loss": 0.5554,
      "step": 1300
    },
    {
      "epoch": 1.1306532663316582,
      "grad_norm": 4.688488006591797,
      "learning_rate": 6.735000000000001e-06,
      "loss": 0.5676,
      "step": 1350
    },
    {
      "epoch": 1.1725293132328307,
      "grad_norm": 3.0410969257354736,
      "learning_rate": 6.985000000000001e-06,
      "loss": 0.5374,
      "step": 1400
    },
    {
      "epoch": 1.2144053601340032,
      "grad_norm": 6.28264856338501,
      "learning_rate": 7.235000000000001e-06,
      "loss": 0.5275,
      "step": 1450
    },
    {
      "epoch": 1.2562814070351758,
      "grad_norm": 3.7520318031311035,
      "learning_rate": 7.485000000000001e-06,
      "loss": 0.5206,
      "step": 1500
    },
    {
      "epoch": 1.2562814070351758,
      "eval_loss": 0.4592318534851074,
      "eval_runtime": 220.2748,
      "eval_samples_per_second": 38.538,
      "eval_steps_per_second": 4.821,
      "step": 1500
    }
  ],
  "logging_steps": 50,
  "max_steps": 8000,
  "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": 1.4557265976097152e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}