checkpoint-5/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8695652173913043,
  "eval_steps": 2,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.9984,
      "step": 1
    },
    {
      "epoch": 0.17,
      "eval_loss": 1.09337317943573,
      "eval_runtime": 65.1601,
      "eval_samples_per_second": 1.535,
      "eval_steps_per_second": 0.767,
      "step": 1
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.9864,
      "step": 2
    },
    {
      "epoch": 0.35,
      "eval_loss": 1.0603257417678833,
      "eval_runtime": 66.8005,
      "eval_samples_per_second": 1.497,
      "eval_steps_per_second": 0.748,
      "step": 2
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.5e-06,
      "loss": 0.9692,
      "step": 3
    },
    {
      "epoch": 0.7,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.9181,
      "step": 4
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.9132175445556641,
      "eval_runtime": 65.6994,
      "eval_samples_per_second": 1.522,
      "eval_steps_per_second": 0.761,
      "step": 4
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.5e-06,
      "loss": 0.9237,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 5,
  "total_flos": 1.398012660154368e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}