trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 21010,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3.941359043121338,
      "learning_rate": 9.932255753356769e-05,
      "loss": 0.835,
      "step": 2101
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3791838884353638,
      "learning_rate": 9.398649562950904e-05,
      "loss": 0.3022,
      "step": 4202
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8936110734939575,
      "learning_rate": 8.388936474556312e-05,
      "loss": 0.1942,
      "step": 6303
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.16859769821167,
      "learning_rate": 7.012307637728611e-05,
      "loss": 0.1428,
      "step": 8404
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5121099948883057,
      "learning_rate": 5.4170246035912365e-05,
      "loss": 0.1132,
      "step": 10505
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.28275570273399353,
      "learning_rate": 3.776548239300111e-05,
      "loss": 0.0947,
      "step": 12606
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.228614330291748,
      "learning_rate": 2.2693176819034056e-05,
      "loss": 0.0841,
      "step": 14707
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.2584194839000702,
      "learning_rate": 1.0577345507714958e-05,
      "loss": 0.0772,
      "step": 16808
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.3184811472892761,
      "learning_rate": 2.7271895019674864e-06,
      "loss": 0.072,
      "step": 18909
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.37828290462493896,
      "learning_rate": 3.0349967744536246e-11,
      "loss": 0.0686,
      "step": 21010
    }
  ],
  "logging_steps": 500,
  "max_steps": 21010,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1001105328078848e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}