trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6442953020134228,
  "global_step": 4900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "learning_rate": 1.9328859060402687e-05,
      "loss": 1.0712,
      "step": 500
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.865771812080537e-05,
      "loss": 0.7775,
      "step": 1000
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.7986577181208054e-05,
      "loss": 0.7932,
      "step": 1500
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.731543624161074e-05,
      "loss": 0.8078,
      "step": 2000
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.6644295302013425e-05,
      "loss": 0.7557,
      "step": 2500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.4945272207260132,
      "eval_runtime": 65.3538,
      "eval_samples_per_second": 17.122,
      "eval_steps_per_second": 5.707,
      "step": 2980
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.5973154362416107e-05,
      "loss": 0.7255,
      "step": 3000
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.5302013422818792e-05,
      "loss": 0.4163,
      "step": 3500
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.4630872483221479e-05,
      "loss": 0.5084,
      "step": 4000
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.3959731543624163e-05,
      "loss": 0.4503,
      "step": 4500
    }
  ],
  "max_steps": 14900,
  "num_train_epochs": 5,
  "total_flos": 1.070174051569728e+16,
  "trial_name": null,
  "trial_params": null
}