trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.993006993006993,
  "eval_steps": 60,
  "global_step": 214,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.27972027972027974,
      "grad_norm": 7.277930736541748,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 1.217,
      "step": 20
    },
    {
      "epoch": 0.5594405594405595,
      "grad_norm": 0.8069848418235779,
      "learning_rate": 5.333333333333333e-05,
      "loss": 0.7878,
      "step": 40
    },
    {
      "epoch": 0.8391608391608392,
      "grad_norm": 1.3099849224090576,
      "learning_rate": 8e-05,
      "loss": 0.6994,
      "step": 60
    },
    {
      "epoch": 0.8391608391608392,
      "eval_loss": 0.6600586771965027,
      "eval_runtime": 33.5004,
      "eval_samples_per_second": 34.209,
      "eval_steps_per_second": 1.075,
      "step": 60
    },
    {
      "epoch": 1.118881118881119,
      "grad_norm": 0.770585298538208,
      "learning_rate": 0.00010666666666666667,
      "loss": 0.6646,
      "step": 80
    },
    {
      "epoch": 1.3986013986013985,
      "grad_norm": 1.0533007383346558,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.6843,
      "step": 100
    },
    {
      "epoch": 1.6783216783216783,
      "grad_norm": 2.8865811824798584,
      "learning_rate": 0.00016,
      "loss": 0.6851,
      "step": 120
    },
    {
      "epoch": 1.6783216783216783,
      "eval_loss": 0.6946624517440796,
      "eval_runtime": 33.3238,
      "eval_samples_per_second": 34.39,
      "eval_steps_per_second": 1.08,
      "step": 120
    },
    {
      "epoch": 1.958041958041958,
      "grad_norm": 1.449644684791565,
      "learning_rate": 0.0001866666666666667,
      "loss": 0.684,
      "step": 140
    },
    {
      "epoch": 2.237762237762238,
      "grad_norm": 1.2769348621368408,
      "learning_rate": 0.00019642857142857144,
      "loss": 0.6842,
      "step": 160
    },
    {
      "epoch": 2.5174825174825175,
      "grad_norm": 1.5653624534606934,
      "learning_rate": 0.0001892857142857143,
      "loss": 0.6734,
      "step": 180
    },
    {
      "epoch": 2.5174825174825175,
      "eval_loss": 0.6711814999580383,
      "eval_runtime": 33.4337,
      "eval_samples_per_second": 34.277,
      "eval_steps_per_second": 1.077,
      "step": 180
    },
    {
      "epoch": 2.797202797202797,
      "grad_norm": 1.9234168529510498,
      "learning_rate": 0.00018214285714285714,
      "loss": 0.6995,
      "step": 200
    }
  ],
  "logging_steps": 20,
  "max_steps": 710,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7224106858315776.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}