trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 6564,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.22851919561243145,
      "grad_norm": 6.620522975921631,
      "learning_rate": 4.6191346739792815e-05,
      "loss": 2.393,
      "step": 500
    },
    {
      "epoch": 0.4570383912248629,
      "grad_norm": 7.987125873565674,
      "learning_rate": 4.238269347958562e-05,
      "loss": 1.7569,
      "step": 1000
    },
    {
      "epoch": 0.6855575868372943,
      "grad_norm": 7.116205215454102,
      "learning_rate": 3.857404021937843e-05,
      "loss": 1.5739,
      "step": 1500
    },
    {
      "epoch": 0.9140767824497258,
      "grad_norm": 7.844669342041016,
      "learning_rate": 3.476538695917124e-05,
      "loss": 1.4561,
      "step": 2000
    },
    {
      "epoch": 1.1425959780621573,
      "grad_norm": 6.016057968139648,
      "learning_rate": 3.0956733698964044e-05,
      "loss": 1.3647,
      "step": 2500
    },
    {
      "epoch": 1.3711151736745886,
      "grad_norm": 5.737545490264893,
      "learning_rate": 2.7148080438756857e-05,
      "loss": 1.2904,
      "step": 3000
    },
    {
      "epoch": 1.59963436928702,
      "grad_norm": 6.617909908294678,
      "learning_rate": 2.3339427178549666e-05,
      "loss": 1.2466,
      "step": 3500
    },
    {
      "epoch": 1.8281535648994516,
      "grad_norm": 6.847684383392334,
      "learning_rate": 1.9530773918342475e-05,
      "loss": 1.2131,
      "step": 4000
    },
    {
      "epoch": 2.056672760511883,
      "grad_norm": 7.49899959564209,
      "learning_rate": 1.5722120658135284e-05,
      "loss": 1.1537,
      "step": 4500
    },
    {
      "epoch": 2.2851919561243146,
      "grad_norm": 6.446639060974121,
      "learning_rate": 1.1913467397928093e-05,
      "loss": 1.1433,
      "step": 5000
    },
    {
      "epoch": 2.5137111517367456,
      "grad_norm": 6.995218753814697,
      "learning_rate": 8.104814137720902e-06,
      "loss": 1.1077,
      "step": 5500
    },
    {
      "epoch": 2.742230347349177,
      "grad_norm": 5.986495018005371,
      "learning_rate": 4.296160877513712e-06,
      "loss": 1.0895,
      "step": 6000
    },
    {
      "epoch": 2.9707495429616086,
      "grad_norm": 6.013411998748779,
      "learning_rate": 4.875076173065205e-07,
      "loss": 1.0868,
      "step": 6500
    },
    {
      "epoch": 3.0,
      "step": 6564,
      "total_flos": 4.896112037854003e+16,
      "train_loss": 1.372001577048386,
      "train_runtime": 85764.1685,
      "train_samples_per_second": 0.612,
      "train_steps_per_second": 0.077
    }
  ],
  "logging_steps": 500,
  "max_steps": 6564,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.896112037854003e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}