trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3.3184640407562256,
      "learning_rate": 1.8004000000000002e-05,
      "loss": 1.903,
      "step": 500
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.314566135406494,
      "learning_rate": 1.6004e-05,
      "loss": 0.8126,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.398888111114502,
      "learning_rate": 1.4004000000000002e-05,
      "loss": 0.7188,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.308872699737549,
      "learning_rate": 1.2004e-05,
      "loss": 0.6496,
      "step": 2000
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.8485875129699707,
      "learning_rate": 1.0004000000000001e-05,
      "loss": 0.5958,
      "step": 2500
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.999493360519409,
      "learning_rate": 8.004e-06,
      "loss": 0.5524,
      "step": 3000
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.292451858520508,
      "learning_rate": 6.004000000000001e-06,
      "loss": 0.5184,
      "step": 3500
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.01755952835083,
      "learning_rate": 4.004e-06,
      "loss": 0.4901,
      "step": 4000
    },
    {
      "epoch": 9.0,
      "grad_norm": 2.8285794258117676,
      "learning_rate": 2.004e-06,
      "loss": 0.4722,
      "step": 4500
    },
    {
      "epoch": 10.0,
      "grad_norm": 2.7315330505371094,
      "learning_rate": 4e-09,
      "loss": 0.4564,
      "step": 5000
    }
  ],
  "logging_steps": 500,
  "max_steps": 5000,
  "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": 6097364582400000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}