trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.903225806451613,
  "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": 4.990048408508301,
      "learning_rate": 1.9954719225730847e-05,
      "loss": 1.1226,
      "step": 5
    },
    {
      "epoch": 0.43010752688172044,
      "grad_norm": 4.11909294128418,
      "learning_rate": 1.9450008187146685e-05,
      "loss": 0.6465,
      "step": 10
    },
    {
      "epoch": 0.6451612903225806,
      "grad_norm": 2.001328706741333,
      "learning_rate": 1.8412535328311813e-05,
      "loss": 0.5056,
      "step": 15
    },
    {
      "epoch": 0.8602150537634409,
      "grad_norm": 1.5412222146987915,
      "learning_rate": 1.6900790114821122e-05,
      "loss": 0.4228,
      "step": 20
    },
    {
      "epoch": 1.043010752688172,
      "grad_norm": 1.5938291549682617,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.3764,
      "step": 25
    },
    {
      "epoch": 1.2580645161290323,
      "grad_norm": 1.3997321128845215,
      "learning_rate": 1.2817325568414299e-05,
      "loss": 0.3061,
      "step": 30
    },
    {
      "epoch": 1.4731182795698925,
      "grad_norm": 1.2927430868148804,
      "learning_rate": 1.0475819158237426e-05,
      "loss": 0.2927,
      "step": 35
    },
    {
      "epoch": 1.6881720430107527,
      "grad_norm": 1.1591598987579346,
      "learning_rate": 8.107487556395902e-06,
      "loss": 0.2849,
      "step": 40
    },
    {
      "epoch": 1.903225806451613,
      "grad_norm": 1.0788626670837402,
      "learning_rate": 5.845849869981137e-06,
      "loss": 0.2609,
      "step": 45
    },
    {
      "epoch": 2.086021505376344,
      "grad_norm": 1.1503716707229614,
      "learning_rate": 3.818410137793947e-06,
      "loss": 0.2433,
      "step": 50
    },
    {
      "epoch": 2.3010752688172045,
      "grad_norm": 1.0680569410324097,
      "learning_rate": 2.1394690525721275e-06,
      "loss": 0.2139,
      "step": 55
    },
    {
      "epoch": 2.5161290322580645,
      "grad_norm": 1.1041356325149536,
      "learning_rate": 9.036800464548157e-07,
      "loss": 0.2041,
      "step": 60
    },
    {
      "epoch": 2.731182795698925,
      "grad_norm": 0.9912173748016357,
      "learning_rate": 1.8071302737293294e-07,
      "loss": 0.2012,
      "step": 65
    },
    {
      "epoch": 2.903225806451613,
      "step": 69,
      "total_flos": 2.582932209061069e+16,
      "train_loss": 0.38006642072097113,
      "train_runtime": 699.6666,
      "train_samples_per_second": 6.359,
      "train_steps_per_second": 0.099
    }
  ],
  "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": 2.582932209061069e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}