trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.999965170629088,
  "global_step": 13458,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 2.6686,
      "step": 500
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 2.3316,
      "step": 1000
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 2.2489,
      "step": 1500
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 2.2069,
      "step": 2000
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 2.1742,
      "step": 2500
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 2.1453,
      "step": 3000
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 2.1187,
      "step": 3500
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 2.103,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 2.0829,
      "step": 4500
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 2.0455,
      "step": 5000
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 2.0346,
      "step": 5500
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 2.0186,
      "step": 6000
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 2.0114,
      "step": 6500
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001,
      "loss": 2.0048,
      "step": 7000
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0001,
      "loss": 1.9923,
      "step": 7500
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001,
      "loss": 1.9883,
      "step": 8000
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0001,
      "loss": 1.9821,
      "step": 8500
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0001,
      "loss": 1.9678,
      "step": 9000
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0001,
      "loss": 1.9427,
      "step": 9500
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0001,
      "loss": 1.938,
      "step": 10000
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0001,
      "loss": 1.9309,
      "step": 10500
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0001,
      "loss": 1.9198,
      "step": 11000
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0001,
      "loss": 1.9246,
      "step": 11500
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001,
      "loss": 1.916,
      "step": 12000
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0001,
      "loss": 1.9193,
      "step": 12500
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0001,
      "loss": 1.9114,
      "step": 13000
    },
    {
      "epoch": 3.0,
      "step": 13458,
      "total_flos": 7.160049636932506e+17,
      "train_loss": 2.0534920626541253,
      "train_runtime": 109961.7964,
      "train_samples_per_second": 7.833,
      "train_steps_per_second": 0.122
    }
  ],
  "max_steps": 13458,
  "num_train_epochs": 3,
  "total_flos": 7.160049636932506e+17,
  "trial_name": null,
  "trial_params": null
}