trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9690721649484537,
  "eval_steps": 500,
  "global_step": 18,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9896907216494846,
      "eval_loss": 1.1644935607910156,
      "eval_runtime": 22.1229,
      "eval_samples_per_second": 7.368,
      "eval_steps_per_second": 0.271,
      "step": 6
    },
    {
      "epoch": 1.6494845360824741,
      "grad_norm": 7.7773427585616135,
      "learning_rate": 5e-06,
      "loss": 1.2224,
      "step": 10
    },
    {
      "epoch": 1.9793814432989691,
      "eval_loss": 1.1061632633209229,
      "eval_runtime": 21.7651,
      "eval_samples_per_second": 7.489,
      "eval_steps_per_second": 0.276,
      "step": 12
    },
    {
      "epoch": 2.9690721649484537,
      "eval_loss": 1.067095398902893,
      "eval_runtime": 21.5996,
      "eval_samples_per_second": 7.546,
      "eval_steps_per_second": 0.278,
      "step": 18
    },
    {
      "epoch": 2.9690721649484537,
      "step": 18,
      "total_flos": 30045980590080.0,
      "train_loss": 1.1581032541063097,
      "train_runtime": 3196.5527,
      "train_samples_per_second": 2.903,
      "train_steps_per_second": 0.006
    }
  ],
  "logging_steps": 10,
  "max_steps": 18,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 30045980590080.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}