trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.851063829787234,
  "eval_steps": 500,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1702127659574468,
      "grad_norm": 1.0480449041222901,
      "learning_rate": 4.980286753286196e-06,
      "loss": 1.1367,
      "step": 1
    },
    {
      "epoch": 0.851063829787234,
      "grad_norm": 0.7218518716452021,
      "learning_rate": 4.522542485937369e-06,
      "loss": 0.7212,
      "step": 5
    },
    {
      "epoch": 1.851063829787234,
      "grad_norm": 0.48974754683120497,
      "learning_rate": 3.272542485937369e-06,
      "loss": 0.5486,
      "step": 10
    },
    {
      "epoch": 2.851063829787234,
      "grad_norm": 0.05989476827451271,
      "learning_rate": 1.7274575140626318e-06,
      "loss": 0.4988,
      "step": 15
    },
    {
      "epoch": 3.851063829787234,
      "grad_norm": 0.04162142765751137,
      "learning_rate": 4.774575140626317e-07,
      "loss": 0.4812,
      "step": 20
    },
    {
      "epoch": 4.851063829787234,
      "grad_norm": 0.03625801845951793,
      "learning_rate": 0.0,
      "loss": 0.4803,
      "step": 25
    },
    {
      "epoch": 4.851063829787234,
      "step": 25,
      "total_flos": 1.70668694534357e+17,
      "train_loss": 0.5626590156555176,
      "train_runtime": 583.7651,
      "train_samples_per_second": 6.407,
      "train_steps_per_second": 0.043
    }
  ],
  "logging_steps": 5,
  "max_steps": 25,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.70668694534357e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}