{
  "best_metric": 0.41453465819358826,
  "best_model_checkpoint": "mikhail-panzo/fil_b128_le3_s4000/checkpoint-500",
  "epoch": 88.88888888888889,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.444444444444445,
      "grad_norm": 1.5276468992233276,
      "learning_rate": 2.5e-05,
      "loss": 0.7204,
      "step": 50
    },
    {
      "epoch": 8.88888888888889,
      "grad_norm": 1.4177956581115723,
      "learning_rate": 5e-05,
      "loss": 0.5494,
      "step": 100
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 3.0381362438201904,
      "learning_rate": 7.5e-05,
      "loss": 0.4948,
      "step": 150
    },
    {
      "epoch": 17.77777777777778,
      "grad_norm": 1.7661292552947998,
      "learning_rate": 0.0001,
      "loss": 0.4734,
      "step": 200
    },
    {
      "epoch": 22.22222222222222,
      "grad_norm": 1.4447826147079468,
      "learning_rate": 0.000125,
      "loss": 0.473,
      "step": 250
    },
    {
      "epoch": 26.666666666666668,
      "grad_norm": 2.745311975479126,
      "learning_rate": 0.00015,
      "loss": 0.464,
      "step": 300
    },
    {
      "epoch": 31.11111111111111,
      "grad_norm": 2.4395411014556885,
      "learning_rate": 0.000175,
      "loss": 0.4593,
      "step": 350
    },
    {
      "epoch": 35.55555555555556,
      "grad_norm": 2.1870596408843994,
      "learning_rate": 0.0002,
      "loss": 0.4452,
      "step": 400
    },
    {
      "epoch": 40.0,
      "grad_norm": 2.9709341526031494,
      "learning_rate": 0.00022500000000000002,
      "loss": 0.4404,
      "step": 450
    },
    {
      "epoch": 44.44444444444444,
      "grad_norm": 3.981621265411377,
      "learning_rate": 0.00025,
      "loss": 0.4367,
      "step": 500
    },
    {
      "epoch": 44.44444444444444,
      "eval_loss": 0.41453465819358826,
      "eval_runtime": 9.8143,
      "eval_samples_per_second": 16.201,
      "eval_steps_per_second": 2.038,
      "step": 500
    },
    {
      "epoch": 48.888888888888886,
      "grad_norm": 9.63549518585205,
      "learning_rate": 0.000275,
      "loss": 0.4377,
      "step": 550
    },
    {
      "epoch": 53.333333333333336,
      "grad_norm": 3.0573575496673584,
      "learning_rate": 0.0003,
      "loss": 0.4377,
      "step": 600
    },
    {
      "epoch": 57.77777777777778,
      "grad_norm": 2.144299268722534,
      "learning_rate": 0.00032500000000000004,
      "loss": 0.4342,
      "step": 650
    },
    {
      "epoch": 62.22222222222222,
      "grad_norm": 3.523078203201294,
      "learning_rate": 0.00035,
      "loss": 0.4375,
      "step": 700
    },
    {
      "epoch": 66.66666666666667,
      "grad_norm": 3.0409202575683594,
      "learning_rate": 0.000375,
      "loss": 0.4382,
      "step": 750
    },
    {
      "epoch": 71.11111111111111,
      "grad_norm": 7.679955959320068,
      "learning_rate": 0.0004,
      "loss": 0.4435,
      "step": 800
    },
    {
      "epoch": 75.55555555555556,
      "grad_norm": 4.8538618087768555,
      "learning_rate": 0.000425,
      "loss": 0.4313,
      "step": 850
    },
    {
      "epoch": 80.0,
      "grad_norm": 3.889315128326416,
      "learning_rate": 0.00045000000000000004,
      "loss": 0.4242,
      "step": 900
    },
    {
      "epoch": 84.44444444444444,
      "grad_norm": 9.169000625610352,
      "learning_rate": 0.000475,
      "loss": 0.4703,
      "step": 950
    },
    {
      "epoch": 88.88888888888889,
      "grad_norm": 3.6193642616271973,
      "learning_rate": 0.0005,
      "loss": 0.4709,
      "step": 1000
    },
    {
      "epoch": 88.88888888888889,
      "eval_loss": 0.4963502883911133,
      "eval_runtime": 9.099,
      "eval_samples_per_second": 17.474,
      "eval_steps_per_second": 2.198,
      "step": 1000
    }
  ],
  "logging_steps": 50,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 364,
  "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": 2.7514700481606624e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}