temp/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "global_step": 220,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.91,
      "learning_rate": 0.0009949107209404665,
      "loss": 4.8778,
      "step": 10
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0009797464868072487,
      "loss": 4.3468,
      "step": 20
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0009548159976772592,
      "loss": 4.0995,
      "step": 30
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.0009206267664155906,
      "loss": 3.6941,
      "step": 40
    },
    {
      "epoch": 4.55,
      "learning_rate": 0.0008825117959999116,
      "loss": 3.3691,
      "step": 50
    },
    {
      "epoch": 5.45,
      "learning_rate": 0.0008327928391111841,
      "loss": 2.969,
      "step": 60
    },
    {
      "epoch": 6.36,
      "learning_rate": 0.0007762991797134513,
      "loss": 2.4856,
      "step": 70
    },
    {
      "epoch": 7.27,
      "learning_rate": 0.0007141808657977907,
      "loss": 2.285,
      "step": 80
    },
    {
      "epoch": 8.18,
      "learning_rate": 0.0006477024471011001,
      "loss": 1.8686,
      "step": 90
    },
    {
      "epoch": 9.09,
      "learning_rate": 0.0005782172325201155,
      "loss": 1.5859,
      "step": 100
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0005071397406448937,
      "loss": 1.2427,
      "step": 110
    },
    {
      "epoch": 10.91,
      "learning_rate": 0.0004359169042394536,
      "loss": 1.0274,
      "step": 120
    },
    {
      "epoch": 11.82,
      "learning_rate": 0.0003659986148633107,
      "loss": 0.8537,
      "step": 130
    },
    {
      "epoch": 12.73,
      "learning_rate": 0.0002988082072604661,
      "loss": 0.6993,
      "step": 140
    },
    {
      "epoch": 13.64,
      "learning_rate": 0.00023571348436857904,
      "loss": 0.4831,
      "step": 150
    },
    {
      "epoch": 14.55,
      "learning_rate": 0.00017799887279557237,
      "loss": 0.4156,
      "step": 160
    },
    {
      "epoch": 15.45,
      "learning_rate": 0.00012683927559787656,
      "loss": 0.3635,
      "step": 170
    },
    {
      "epoch": 16.36,
      "learning_rate": 8.327615464234128e-05,
      "loss": 0.3198,
      "step": 180
    },
    {
      "epoch": 17.27,
      "learning_rate": 4.819632944595414e-05,
      "loss": 0.2574,
      "step": 190
    },
    {
      "epoch": 18.18,
      "learning_rate": 2.2313924087851657e-05,
      "loss": 0.2336,
      "step": 200
    },
    {
      "epoch": 19.09,
      "learning_rate": 6.15582970243117e-06,
      "loss": 0.2314,
      "step": 210
    },
    {
      "epoch": 20.0,
      "learning_rate": 5.0978495283349194e-08,
      "loss": 0.215,
      "step": 220
    },
    {
      "epoch": 20.0,
      "step": 220,
      "total_flos": 5.476535779894886e+16,
      "train_loss": 1.7238198778846048,
      "train_runtime": 2069.8284,
      "train_samples_per_second": 1.672,
      "train_steps_per_second": 0.106
    }
  ],
  "max_steps": 220,
  "num_train_epochs": 20,
  "total_flos": 5.476535779894886e+16,
  "trial_name": null,
  "trial_params": null
}