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": 0.4083499610424042,
      "learning_rate": 2e-05,
      "loss": 0.2679,
      "step": 1
    },
    {
      "epoch": 0.0004253056884635832,
      "eval_loss": 0.4911474883556366,
      "eval_runtime": 56.6823,
      "eval_samples_per_second": 17.466,
      "eval_steps_per_second": 8.733,
      "step": 1
    },
    {
      "epoch": 0.0008506113769271664,
      "grad_norm": 0.7589439749717712,
      "learning_rate": 4e-05,
      "loss": 0.5251,
      "step": 2
    },
    {
      "epoch": 0.0012759170653907496,
      "grad_norm": 0.6894279718399048,
      "learning_rate": 6e-05,
      "loss": 0.4982,
      "step": 3
    },
    {
      "epoch": 0.0012759170653907496,
      "eval_loss": 0.4878028631210327,
      "eval_runtime": 55.2338,
      "eval_samples_per_second": 17.924,
      "eval_steps_per_second": 8.962,
      "step": 3
    },
    {
      "epoch": 0.0017012227538543328,
      "grad_norm": 0.3619888424873352,
      "learning_rate": 8e-05,
      "loss": 0.2692,
      "step": 4
    },
    {
      "epoch": 0.002126528442317916,
      "grad_norm": 0.45285743474960327,
      "learning_rate": 0.0001,
      "loss": 0.3688,
      "step": 5
    },
    {
      "epoch": 0.002551834130781499,
      "grad_norm": 0.8966401815414429,
      "learning_rate": 0.00012,
      "loss": 0.592,
      "step": 6
    },
    {
      "epoch": 0.002551834130781499,
      "eval_loss": 0.4417019188404083,
      "eval_runtime": 55.3676,
      "eval_samples_per_second": 17.88,
      "eval_steps_per_second": 8.94,
      "step": 6
    },
    {
      "epoch": 0.0029771398192450825,
      "grad_norm": 0.5397375822067261,
      "learning_rate": 0.00014,
      "loss": 0.3469,
      "step": 7
    },
    {
      "epoch": 0.0034024455077086655,
      "grad_norm": 0.4392431676387787,
      "learning_rate": 0.00016,
      "loss": 0.2646,
      "step": 8
    },
    {
      "epoch": 0.003827751196172249,
      "grad_norm": 0.9264920353889465,
      "learning_rate": 0.00018,
      "loss": 0.4984,
      "step": 9
    },
    {
      "epoch": 0.003827751196172249,
      "eval_loss": 0.30864137411117554,
      "eval_runtime": 55.347,
      "eval_samples_per_second": 17.887,
      "eval_steps_per_second": 8.944,
      "step": 9
    },
    {
      "epoch": 0.004253056884635832,
      "grad_norm": 0.39683446288108826,
      "learning_rate": 0.0002,
      "loss": 0.2201,
      "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": 3115469494026240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}