trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.967741935483871,
  "eval_steps": 500,
  "global_step": 69,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21505376344086022,
      "grad_norm": 5.693103313446045,
      "learning_rate": 1.9954719225730847e-05,
      "loss": 1.1669,
      "step": 5
    },
    {
      "epoch": 0.43010752688172044,
      "grad_norm": 1.963355302810669,
      "learning_rate": 1.9450008187146685e-05,
      "loss": 0.6522,
      "step": 10
    },
    {
      "epoch": 0.6451612903225806,
      "grad_norm": 1.6932282447814941,
      "learning_rate": 1.8412535328311813e-05,
      "loss": 0.4912,
      "step": 15
    },
    {
      "epoch": 0.8602150537634409,
      "grad_norm": 1.487468957901001,
      "learning_rate": 1.6900790114821122e-05,
      "loss": 0.415,
      "step": 20
    },
    {
      "epoch": 1.075268817204301,
      "grad_norm": 1.5301791429519653,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.3687,
      "step": 25
    },
    {
      "epoch": 1.2903225806451613,
      "grad_norm": 1.3023558855056763,
      "learning_rate": 1.2817325568414299e-05,
      "loss": 0.2981,
      "step": 30
    },
    {
      "epoch": 1.5053763440860215,
      "grad_norm": 1.2679604291915894,
      "learning_rate": 1.0475819158237426e-05,
      "loss": 0.2836,
      "step": 35
    },
    {
      "epoch": 1.7204301075268817,
      "grad_norm": 1.180540919303894,
      "learning_rate": 8.107487556395902e-06,
      "loss": 0.2772,
      "step": 40
    },
    {
      "epoch": 1.935483870967742,
      "grad_norm": 1.1995508670806885,
      "learning_rate": 5.845849869981137e-06,
      "loss": 0.268,
      "step": 45
    },
    {
      "epoch": 2.150537634408602,
      "grad_norm": 1.1225168704986572,
      "learning_rate": 3.818410137793947e-06,
      "loss": 0.2306,
      "step": 50
    },
    {
      "epoch": 2.3655913978494625,
      "grad_norm": 1.166407823562622,
      "learning_rate": 2.1394690525721275e-06,
      "loss": 0.2085,
      "step": 55
    },
    {
      "epoch": 2.5806451612903225,
      "grad_norm": 1.157459020614624,
      "learning_rate": 9.036800464548157e-07,
      "loss": 0.2009,
      "step": 60
    },
    {
      "epoch": 2.795698924731183,
      "grad_norm": 1.1147010326385498,
      "learning_rate": 1.8071302737293294e-07,
      "loss": 0.2012,
      "step": 65
    },
    {
      "epoch": 2.967741935483871,
      "step": 69,
      "total_flos": 2.640330688875725e+16,
      "train_loss": 0.37769337063250336,
      "train_runtime": 647.3014,
      "train_samples_per_second": 6.873,
      "train_steps_per_second": 0.107
    }
  ],
  "logging_steps": 5,
  "max_steps": 69,
  "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": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.640330688875725e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}