last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.7648864333947207,
  "eval_steps": 500,
  "global_step": 11500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 0.0009846531614487416,
      "loss": 0.5762,
      "step": 500
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0009693063228974831,
      "loss": 0.6042,
      "step": 1000
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0009539594843462247,
      "loss": 0.6024,
      "step": 1500
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0009386126457949663,
      "loss": 0.6057,
      "step": 2000
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0009232658072437078,
      "loss": 0.6143,
      "step": 2500
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0009079189686924494,
      "loss": 0.6081,
      "step": 3000
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.000892572130141191,
      "loss": 0.611,
      "step": 3500
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0008772252915899325,
      "loss": 0.6104,
      "step": 4000
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0008619091467157766,
      "loss": 0.6154,
      "step": 4500
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0008465623081645181,
      "loss": 0.6151,
      "step": 5000
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0008312154696132597,
      "loss": 0.611,
      "step": 5500
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0008158686310620012,
      "loss": 0.617,
      "step": 6000
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0008005524861878454,
      "loss": 0.6089,
      "step": 6500
    },
    {
      "epoch": 1.0,
      "eval_bleu": 3.6822,
      "eval_gen_len": 18.235,
      "eval_loss": 0.506125271320343,
      "eval_runtime": 27.1471,
      "eval_samples_per_second": 36.836,
      "eval_steps_per_second": 2.321,
      "step": 6516
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0007852056476365869,
      "loss": 0.5657,
      "step": 7000
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0007698588090853285,
      "loss": 0.5623,
      "step": 7500
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0007545119705340699,
      "loss": 0.5669,
      "step": 8000
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.000739195825659914,
      "loss": 0.5793,
      "step": 8500
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0007238489871086557,
      "loss": 0.5743,
      "step": 9000
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0007085328422344997,
      "loss": 0.5722,
      "step": 9500
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006931860036832413,
      "loss": 0.5749,
      "step": 10000
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006778391651319828,
      "loss": 0.5776,
      "step": 10500
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006624923265807244,
      "loss": 0.5694,
      "step": 11000
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006471454880294659,
      "loss": 0.5716,
      "step": 11500
    }
  ],
  "logging_steps": 500,
  "max_steps": 32580,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 6096687148302336.0,
  "trial_name": null,
  "trial_params": null
}