trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 106,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18867924528301888,
      "grad_norm": 0.7883548026054572,
      "learning_rate": 9.9778098230154e-06,
      "loss": 0.3824,
      "step": 10
    },
    {
      "epoch": 0.37735849056603776,
      "grad_norm": 0.6278392436846756,
      "learning_rate": 9.588773128419907e-06,
      "loss": 0.2385,
      "step": 20
    },
    {
      "epoch": 0.5660377358490566,
      "grad_norm": 0.4999280479126666,
      "learning_rate": 8.750555348152299e-06,
      "loss": 0.2252,
      "step": 30
    },
    {
      "epoch": 0.7547169811320755,
      "grad_norm": 0.534914138022231,
      "learning_rate": 7.545207078751858e-06,
      "loss": 0.2205,
      "step": 40
    },
    {
      "epoch": 0.9433962264150944,
      "grad_norm": 0.4906193601210189,
      "learning_rate": 6.090716206982714e-06,
      "loss": 0.205,
      "step": 50
    },
    {
      "epoch": 1.1320754716981132,
      "grad_norm": 0.4362201656860912,
      "learning_rate": 4.529458433407429e-06,
      "loss": 0.1606,
      "step": 60
    },
    {
      "epoch": 1.320754716981132,
      "grad_norm": 0.5102083992832466,
      "learning_rate": 3.0142605468260976e-06,
      "loss": 0.1417,
      "step": 70
    },
    {
      "epoch": 1.509433962264151,
      "grad_norm": 0.4576386664144016,
      "learning_rate": 1.6934406733817417e-06,
      "loss": 0.1389,
      "step": 80
    },
    {
      "epoch": 1.6981132075471699,
      "grad_norm": 0.40590022089819333,
      "learning_rate": 6.962898649802824e-07,
      "loss": 0.1362,
      "step": 90
    },
    {
      "epoch": 1.8867924528301887,
      "grad_norm": 0.49013202054176347,
      "learning_rate": 1.2041619030626283e-07,
      "loss": 0.1341,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 106,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.7622188542590976e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}