trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9844961240310077,
  "eval_steps": 500,
  "global_step": 192,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15503875968992248,
      "grad_norm": 1.7575604071433997,
      "learning_rate": 2e-06,
      "loss": 0.7202,
      "step": 10
    },
    {
      "epoch": 0.31007751937984496,
      "grad_norm": 1.0487134821643405,
      "learning_rate": 2e-06,
      "loss": 0.6413,
      "step": 20
    },
    {
      "epoch": 0.46511627906976744,
      "grad_norm": 1.3935531170733104,
      "learning_rate": 2e-06,
      "loss": 0.6142,
      "step": 30
    },
    {
      "epoch": 0.6201550387596899,
      "grad_norm": 1.9050148061428194,
      "learning_rate": 2e-06,
      "loss": 0.6023,
      "step": 40
    },
    {
      "epoch": 0.7751937984496124,
      "grad_norm": 2.222047793133745,
      "learning_rate": 2e-06,
      "loss": 0.5917,
      "step": 50
    },
    {
      "epoch": 0.9302325581395349,
      "grad_norm": 3.019087028888769,
      "learning_rate": 2e-06,
      "loss": 0.5887,
      "step": 60
    },
    {
      "epoch": 1.0930232558139534,
      "grad_norm": 1.7976950197959425,
      "learning_rate": 2e-06,
      "loss": 0.5704,
      "step": 70
    },
    {
      "epoch": 1.248062015503876,
      "grad_norm": 1.8533155798746592,
      "learning_rate": 2e-06,
      "loss": 0.5637,
      "step": 80
    },
    {
      "epoch": 1.4031007751937985,
      "grad_norm": 1.8482551483947047,
      "learning_rate": 2e-06,
      "loss": 0.5642,
      "step": 90
    },
    {
      "epoch": 1.558139534883721,
      "grad_norm": 1.7854116197084653,
      "learning_rate": 2e-06,
      "loss": 0.5624,
      "step": 100
    },
    {
      "epoch": 1.7131782945736433,
      "grad_norm": 1.9103042367556917,
      "learning_rate": 2e-06,
      "loss": 0.5599,
      "step": 110
    },
    {
      "epoch": 1.8682170542635659,
      "grad_norm": 1.6999563845156989,
      "learning_rate": 2e-06,
      "loss": 0.559,
      "step": 120
    },
    {
      "epoch": 2.0232558139534884,
      "grad_norm": 2.049997722312676,
      "learning_rate": 2e-06,
      "loss": 0.5561,
      "step": 130
    },
    {
      "epoch": 2.1782945736434107,
      "grad_norm": 1.7479962190862357,
      "learning_rate": 2e-06,
      "loss": 0.5429,
      "step": 140
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 1.9114226491430213,
      "learning_rate": 2e-06,
      "loss": 0.5412,
      "step": 150
    },
    {
      "epoch": 2.488372093023256,
      "grad_norm": 1.6272735310603161,
      "learning_rate": 2e-06,
      "loss": 0.5415,
      "step": 160
    },
    {
      "epoch": 2.6434108527131785,
      "grad_norm": 1.3870264613858654,
      "learning_rate": 2e-06,
      "loss": 0.5398,
      "step": 170
    },
    {
      "epoch": 2.798449612403101,
      "grad_norm": 1.3363046393730011,
      "learning_rate": 2e-06,
      "loss": 0.5388,
      "step": 180
    },
    {
      "epoch": 2.953488372093023,
      "grad_norm": 1.2806111013768993,
      "learning_rate": 2e-06,
      "loss": 0.5387,
      "step": 190
    },
    {
      "epoch": 2.9844961240310077,
      "step": 192,
      "total_flos": 2566157060014080.0,
      "train_loss": 0.3678056833644708,
      "train_runtime": 2871.1751,
      "train_samples_per_second": 275.814,
      "train_steps_per_second": 0.067
    }
  ],
  "logging_steps": 10,
  "max_steps": 192,
  "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": 2566157060014080.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}