last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004253056884635832,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004253056884635832,
      "grad_norm": 1.1514711380004883,
      "learning_rate": 2e-05,
      "loss": 1.7191,
      "step": 1
    },
    {
      "epoch": 0.0004253056884635832,
      "eval_loss": 2.6664950847625732,
      "eval_runtime": 37.5357,
      "eval_samples_per_second": 26.375,
      "eval_steps_per_second": 13.187,
      "step": 1
    },
    {
      "epoch": 0.0008506113769271664,
      "grad_norm": 9.661757469177246,
      "learning_rate": 4e-05,
      "loss": 2.4374,
      "step": 2
    },
    {
      "epoch": 0.0012759170653907496,
      "grad_norm": 6.818915843963623,
      "learning_rate": 6e-05,
      "loss": 2.1022,
      "step": 3
    },
    {
      "epoch": 0.0012759170653907496,
      "eval_loss": 2.6000125408172607,
      "eval_runtime": 36.5523,
      "eval_samples_per_second": 27.084,
      "eval_steps_per_second": 13.542,
      "step": 3
    },
    {
      "epoch": 0.0017012227538543328,
      "grad_norm": 9.478006362915039,
      "learning_rate": 8e-05,
      "loss": 2.5446,
      "step": 4
    },
    {
      "epoch": 0.002126528442317916,
      "grad_norm": 1.3644733428955078,
      "learning_rate": 0.0001,
      "loss": 1.7551,
      "step": 5
    },
    {
      "epoch": 0.002551834130781499,
      "grad_norm": 3.261920690536499,
      "learning_rate": 0.00012,
      "loss": 1.9667,
      "step": 6
    },
    {
      "epoch": 0.002551834130781499,
      "eval_loss": 2.2514076232910156,
      "eval_runtime": 36.7694,
      "eval_samples_per_second": 26.925,
      "eval_steps_per_second": 13.462,
      "step": 6
    },
    {
      "epoch": 0.0029771398192450825,
      "grad_norm": 1.7460393905639648,
      "learning_rate": 0.00014,
      "loss": 2.2021,
      "step": 7
    },
    {
      "epoch": 0.0034024455077086655,
      "grad_norm": 1.9718475341796875,
      "learning_rate": 0.00016,
      "loss": 2.169,
      "step": 8
    },
    {
      "epoch": 0.003827751196172249,
      "grad_norm": 2.024656295776367,
      "learning_rate": 0.00018,
      "loss": 2.3258,
      "step": 9
    },
    {
      "epoch": 0.003827751196172249,
      "eval_loss": 2.043501377105713,
      "eval_runtime": 36.5039,
      "eval_samples_per_second": 27.12,
      "eval_steps_per_second": 13.56,
      "step": 9
    },
    {
      "epoch": 0.004253056884635832,
      "grad_norm": 1.4730572700500488,
      "learning_rate": 0.0002,
      "loss": 1.9354,
      "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": 511583956500480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}