trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.996168582375478,
  "global_step": 650,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.77,
      "learning_rate": 4.615384615384616e-05,
      "loss": 9.0082,
      "step": 50
    },
    {
      "epoch": 0.77,
      "eval_loss": 8.411425590515137,
      "eval_runtime": 12.7191,
      "eval_samples_per_second": 45.601,
      "eval_steps_per_second": 2.909,
      "step": 50
    },
    {
      "epoch": 1.54,
      "learning_rate": 4.230769230769231e-05,
      "loss": 8.0104,
      "step": 100
    },
    {
      "epoch": 1.54,
      "eval_loss": 7.781517505645752,
      "eval_runtime": 12.722,
      "eval_samples_per_second": 45.59,
      "eval_steps_per_second": 2.908,
      "step": 100
    },
    {
      "epoch": 2.31,
      "learning_rate": 3.846153846153846e-05,
      "loss": 7.6542,
      "step": 150
    },
    {
      "epoch": 2.31,
      "eval_loss": 7.551076889038086,
      "eval_runtime": 12.717,
      "eval_samples_per_second": 45.608,
      "eval_steps_per_second": 2.909,
      "step": 150
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.461538461538462e-05,
      "loss": 7.4951,
      "step": 200
    },
    {
      "epoch": 3.08,
      "eval_loss": 7.532299995422363,
      "eval_runtime": 12.7257,
      "eval_samples_per_second": 45.577,
      "eval_steps_per_second": 2.908,
      "step": 200
    },
    {
      "epoch": 3.84,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 7.3489,
      "step": 250
    },
    {
      "epoch": 3.84,
      "eval_loss": 7.389176845550537,
      "eval_runtime": 12.723,
      "eval_samples_per_second": 45.587,
      "eval_steps_per_second": 2.908,
      "step": 250
    },
    {
      "epoch": 4.61,
      "learning_rate": 2.6923076923076923e-05,
      "loss": 7.349,
      "step": 300
    },
    {
      "epoch": 4.61,
      "eval_loss": 7.34345006942749,
      "eval_runtime": 12.7274,
      "eval_samples_per_second": 45.571,
      "eval_steps_per_second": 2.907,
      "step": 300
    },
    {
      "epoch": 5.38,
      "learning_rate": 2.307692307692308e-05,
      "loss": 7.2625,
      "step": 350
    },
    {
      "epoch": 5.38,
      "eval_loss": 7.265749454498291,
      "eval_runtime": 12.7257,
      "eval_samples_per_second": 45.577,
      "eval_steps_per_second": 2.907,
      "step": 350
    },
    {
      "epoch": 6.15,
      "learning_rate": 1.923076923076923e-05,
      "loss": 7.1982,
      "step": 400
    },
    {
      "epoch": 6.15,
      "eval_loss": 7.229588985443115,
      "eval_runtime": 12.7294,
      "eval_samples_per_second": 45.564,
      "eval_steps_per_second": 2.907,
      "step": 400
    },
    {
      "epoch": 6.92,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 7.1324,
      "step": 450
    },
    {
      "epoch": 6.92,
      "eval_loss": 7.157928466796875,
      "eval_runtime": 12.7187,
      "eval_samples_per_second": 45.602,
      "eval_steps_per_second": 2.909,
      "step": 450
    },
    {
      "epoch": 7.69,
      "learning_rate": 1.153846153846154e-05,
      "loss": 7.1505,
      "step": 500
    },
    {
      "epoch": 7.69,
      "eval_loss": 7.151326656341553,
      "eval_runtime": 12.7324,
      "eval_samples_per_second": 45.553,
      "eval_steps_per_second": 2.906,
      "step": 500
    },
    {
      "epoch": 8.46,
      "learning_rate": 7.692307692307694e-06,
      "loss": 7.1079,
      "step": 550
    },
    {
      "epoch": 8.46,
      "eval_loss": 7.1157402992248535,
      "eval_runtime": 12.7267,
      "eval_samples_per_second": 45.573,
      "eval_steps_per_second": 2.907,
      "step": 550
    },
    {
      "epoch": 9.23,
      "learning_rate": 3.846153846153847e-06,
      "loss": 7.1039,
      "step": 600
    },
    {
      "epoch": 9.23,
      "eval_loss": 7.14491605758667,
      "eval_runtime": 12.7279,
      "eval_samples_per_second": 45.569,
      "eval_steps_per_second": 2.907,
      "step": 600
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0,
      "loss": 7.046,
      "step": 650
    },
    {
      "epoch": 10.0,
      "eval_loss": 7.087325096130371,
      "eval_runtime": 12.7263,
      "eval_samples_per_second": 45.575,
      "eval_steps_per_second": 2.907,
      "step": 650
    }
  ],
  "max_steps": 650,
  "num_train_epochs": 10,
  "total_flos": 1.3722182915328e+16,
  "trial_name": null,
  "trial_params": null
}