{
  "best_metric": 0.4247289001941681,
  "best_model_checkpoint": "mikhail-panzo/ceb_b64_le5_s4000/checkpoint-1000",
  "epoch": 39.603960396039604,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.9801980198019802,
      "grad_norm": 2.883972644805908,
      "learning_rate": 2.5000000000000004e-07,
      "loss": 0.8004,
      "step": 50
    },
    {
      "epoch": 3.9603960396039604,
      "grad_norm": 2.5420947074890137,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.77,
      "step": 100
    },
    {
      "epoch": 5.9405940594059405,
      "grad_norm": 1.3930779695510864,
      "learning_rate": 7.5e-07,
      "loss": 0.7525,
      "step": 150
    },
    {
      "epoch": 7.920792079207921,
      "grad_norm": 1.8414520025253296,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.722,
      "step": 200
    },
    {
      "epoch": 9.900990099009901,
      "grad_norm": 2.0665407180786133,
      "learning_rate": 1.25e-06,
      "loss": 0.7076,
      "step": 250
    },
    {
      "epoch": 11.881188118811881,
      "grad_norm": 1.2595587968826294,
      "learning_rate": 1.5e-06,
      "loss": 0.6757,
      "step": 300
    },
    {
      "epoch": 13.861386138613861,
      "grad_norm": 1.4587587118148804,
      "learning_rate": 1.75e-06,
      "loss": 0.6562,
      "step": 350
    },
    {
      "epoch": 15.841584158415841,
      "grad_norm": 2.241546154022217,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.6016,
      "step": 400
    },
    {
      "epoch": 17.821782178217823,
      "grad_norm": 1.4243369102478027,
      "learning_rate": 2.25e-06,
      "loss": 0.5593,
      "step": 450
    },
    {
      "epoch": 19.801980198019802,
      "grad_norm": 1.6889928579330444,
      "learning_rate": 2.5e-06,
      "loss": 0.536,
      "step": 500
    },
    {
      "epoch": 19.801980198019802,
      "eval_loss": 0.4640245735645294,
      "eval_runtime": 6.7454,
      "eval_samples_per_second": 26.685,
      "eval_steps_per_second": 3.41,
      "step": 500
    },
    {
      "epoch": 21.782178217821784,
      "grad_norm": 1.1601781845092773,
      "learning_rate": 2.7500000000000004e-06,
      "loss": 0.5198,
      "step": 550
    },
    {
      "epoch": 23.762376237623762,
      "grad_norm": 1.1969332695007324,
      "learning_rate": 3e-06,
      "loss": 0.5114,
      "step": 600
    },
    {
      "epoch": 25.742574257425744,
      "grad_norm": 1.1022064685821533,
      "learning_rate": 3.2500000000000002e-06,
      "loss": 0.5092,
      "step": 650
    },
    {
      "epoch": 27.722772277227723,
      "grad_norm": 1.2418344020843506,
      "learning_rate": 3.5e-06,
      "loss": 0.4951,
      "step": 700
    },
    {
      "epoch": 29.702970297029704,
      "grad_norm": 1.1083300113677979,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.4911,
      "step": 750
    },
    {
      "epoch": 31.683168316831683,
      "grad_norm": 1.2230323553085327,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.4866,
      "step": 800
    },
    {
      "epoch": 33.663366336633665,
      "grad_norm": 1.1232423782348633,
      "learning_rate": 4.25e-06,
      "loss": 0.4876,
      "step": 850
    },
    {
      "epoch": 35.64356435643565,
      "grad_norm": 0.9751402139663696,
      "learning_rate": 4.5e-06,
      "loss": 0.4772,
      "step": 900
    },
    {
      "epoch": 37.62376237623762,
      "grad_norm": 1.0373197793960571,
      "learning_rate": 4.75e-06,
      "loss": 0.4791,
      "step": 950
    },
    {
      "epoch": 39.603960396039604,
      "grad_norm": 1.0164889097213745,
      "learning_rate": 5e-06,
      "loss": 0.4674,
      "step": 1000
    },
    {
      "epoch": 39.603960396039604,
      "eval_loss": 0.4247289001941681,
      "eval_runtime": 6.5527,
      "eval_samples_per_second": 27.47,
      "eval_steps_per_second": 3.51,
      "step": 1000
    }
  ],
  "logging_steps": 50,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 160,
  "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": 1.0803690034289064e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}