last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00030028677386904494,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.0028677386904494e-05,
      "grad_norm": 0.1895468384027481,
      "learning_rate": 2e-05,
      "loss": 0.3476,
      "step": 1
    },
    {
      "epoch": 3.0028677386904494e-05,
      "eval_loss": 0.3906751573085785,
      "eval_runtime": 1795.9888,
      "eval_samples_per_second": 7.807,
      "eval_steps_per_second": 3.904,
      "step": 1
    },
    {
      "epoch": 6.005735477380899e-05,
      "grad_norm": 0.3096417784690857,
      "learning_rate": 4e-05,
      "loss": 0.5332,
      "step": 2
    },
    {
      "epoch": 9.008603216071348e-05,
      "grad_norm": 0.1562044620513916,
      "learning_rate": 6e-05,
      "loss": 0.2813,
      "step": 3
    },
    {
      "epoch": 9.008603216071348e-05,
      "eval_loss": 0.38901370763778687,
      "eval_runtime": 1876.3804,
      "eval_samples_per_second": 7.473,
      "eval_steps_per_second": 3.736,
      "step": 3
    },
    {
      "epoch": 0.00012011470954761798,
      "grad_norm": 0.2617030143737793,
      "learning_rate": 8e-05,
      "loss": 0.3778,
      "step": 4
    },
    {
      "epoch": 0.00015014338693452247,
      "grad_norm": 0.2616826891899109,
      "learning_rate": 0.0001,
      "loss": 0.3016,
      "step": 5
    },
    {
      "epoch": 0.00018017206432142696,
      "grad_norm": 0.3064740002155304,
      "learning_rate": 0.00012,
      "loss": 0.4085,
      "step": 6
    },
    {
      "epoch": 0.00018017206432142696,
      "eval_loss": 0.359825074672699,
      "eval_runtime": 1107.5181,
      "eval_samples_per_second": 12.661,
      "eval_steps_per_second": 6.33,
      "step": 6
    },
    {
      "epoch": 0.00021020074170833146,
      "grad_norm": 0.24719107151031494,
      "learning_rate": 0.00014,
      "loss": 0.361,
      "step": 7
    },
    {
      "epoch": 0.00024022941909523595,
      "grad_norm": 0.3021886944770813,
      "learning_rate": 0.00016,
      "loss": 0.2602,
      "step": 8
    },
    {
      "epoch": 0.0002702580964821404,
      "grad_norm": 0.5306755304336548,
      "learning_rate": 0.00018,
      "loss": 0.2036,
      "step": 9
    },
    {
      "epoch": 0.0002702580964821404,
      "eval_loss": 0.23427625000476837,
      "eval_runtime": 1094.894,
      "eval_samples_per_second": 12.807,
      "eval_steps_per_second": 6.403,
      "step": 9
    },
    {
      "epoch": 0.00030028677386904494,
      "grad_norm": 0.46173959970474243,
      "learning_rate": 0.0002,
      "loss": 0.3107,
      "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": 5540857302220800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}