checkpoint-1500/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8508224617129893,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.28360748723766305,
      "grad_norm": 4.999021530151367,
      "learning_rate": 1.971639251276234e-05,
      "loss": 6.7525,
      "step": 500
    },
    {
      "epoch": 0.5672149744753261,
      "grad_norm": 8.877035140991211,
      "learning_rate": 1.9432785025524678e-05,
      "loss": 4.8633,
      "step": 1000
    },
    {
      "epoch": 0.8508224617129893,
      "grad_norm": 5.504563808441162,
      "learning_rate": 1.9149177538287012e-05,
      "loss": 4.7848,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 35260,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 3.0466268725248e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}