last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0016410929679166325,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00016410929679166325,
      "grad_norm": 1.91439688205719,
      "learning_rate": 2e-05,
      "loss": 2.5636,
      "step": 1
    },
    {
      "epoch": 0.00016410929679166325,
      "eval_loss": 2.553264856338501,
      "eval_runtime": 392.8391,
      "eval_samples_per_second": 6.532,
      "eval_steps_per_second": 3.266,
      "step": 1
    },
    {
      "epoch": 0.0003282185935833265,
      "grad_norm": 2.276977777481079,
      "learning_rate": 4e-05,
      "loss": 3.1142,
      "step": 2
    },
    {
      "epoch": 0.0004923278903749898,
      "grad_norm": 1.631351113319397,
      "learning_rate": 6e-05,
      "loss": 2.6682,
      "step": 3
    },
    {
      "epoch": 0.0004923278903749898,
      "eval_loss": 2.547764301300049,
      "eval_runtime": 393.137,
      "eval_samples_per_second": 6.527,
      "eval_steps_per_second": 3.263,
      "step": 3
    },
    {
      "epoch": 0.000656437187166653,
      "grad_norm": 1.2931513786315918,
      "learning_rate": 8e-05,
      "loss": 2.0285,
      "step": 4
    },
    {
      "epoch": 0.0008205464839583163,
      "grad_norm": 1.8168163299560547,
      "learning_rate": 0.0001,
      "loss": 2.5827,
      "step": 5
    },
    {
      "epoch": 0.0009846557807499795,
      "grad_norm": 1.7620759010314941,
      "learning_rate": 0.00012,
      "loss": 2.4112,
      "step": 6
    },
    {
      "epoch": 0.0009846557807499795,
      "eval_loss": 2.480220317840576,
      "eval_runtime": 392.327,
      "eval_samples_per_second": 6.54,
      "eval_steps_per_second": 3.27,
      "step": 6
    },
    {
      "epoch": 0.0011487650775416428,
      "grad_norm": 1.6749258041381836,
      "learning_rate": 0.00014,
      "loss": 2.7895,
      "step": 7
    },
    {
      "epoch": 0.001312874374333306,
      "grad_norm": 1.4969555139541626,
      "learning_rate": 0.00016,
      "loss": 2.375,
      "step": 8
    },
    {
      "epoch": 0.0014769836711249693,
      "grad_norm": 1.5681908130645752,
      "learning_rate": 0.00018,
      "loss": 2.414,
      "step": 9
    },
    {
      "epoch": 0.0014769836711249693,
      "eval_loss": 2.3583710193634033,
      "eval_runtime": 391.9968,
      "eval_samples_per_second": 6.546,
      "eval_steps_per_second": 3.273,
      "step": 9
    },
    {
      "epoch": 0.0016410929679166325,
      "grad_norm": 2.345632553100586,
      "learning_rate": 0.0002,
      "loss": 2.4142,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1935445139128320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}