last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.017543859649122806,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0017543859649122807,
      "grad_norm": 3.228858470916748,
      "learning_rate": 2e-05,
      "loss": 2.215,
      "step": 1
    },
    {
      "epoch": 0.0017543859649122807,
      "eval_loss": 2.053115129470825,
      "eval_runtime": 12.1447,
      "eval_samples_per_second": 19.762,
      "eval_steps_per_second": 9.881,
      "step": 1
    },
    {
      "epoch": 0.0035087719298245615,
      "grad_norm": 2.878025531768799,
      "learning_rate": 4e-05,
      "loss": 1.8049,
      "step": 2
    },
    {
      "epoch": 0.005263157894736842,
      "grad_norm": 2.509884834289551,
      "learning_rate": 6e-05,
      "loss": 1.7352,
      "step": 3
    },
    {
      "epoch": 0.005263157894736842,
      "eval_loss": 1.972490668296814,
      "eval_runtime": 12.1634,
      "eval_samples_per_second": 19.731,
      "eval_steps_per_second": 9.866,
      "step": 3
    },
    {
      "epoch": 0.007017543859649123,
      "grad_norm": 2.6294302940368652,
      "learning_rate": 8e-05,
      "loss": 1.866,
      "step": 4
    },
    {
      "epoch": 0.008771929824561403,
      "grad_norm": 3.6487789154052734,
      "learning_rate": 0.0001,
      "loss": 2.1895,
      "step": 5
    },
    {
      "epoch": 0.010526315789473684,
      "grad_norm": 2.667315721511841,
      "learning_rate": 0.00012,
      "loss": 1.4294,
      "step": 6
    },
    {
      "epoch": 0.010526315789473684,
      "eval_loss": 1.2546732425689697,
      "eval_runtime": 12.3056,
      "eval_samples_per_second": 19.503,
      "eval_steps_per_second": 9.752,
      "step": 6
    },
    {
      "epoch": 0.012280701754385965,
      "grad_norm": 2.4955966472625732,
      "learning_rate": 0.00014,
      "loss": 1.223,
      "step": 7
    },
    {
      "epoch": 0.014035087719298246,
      "grad_norm": 1.8106328248977661,
      "learning_rate": 0.00016,
      "loss": 0.9325,
      "step": 8
    },
    {
      "epoch": 0.015789473684210527,
      "grad_norm": 1.708560824394226,
      "learning_rate": 0.00018,
      "loss": 0.8574,
      "step": 9
    },
    {
      "epoch": 0.015789473684210527,
      "eval_loss": 0.8678547739982605,
      "eval_runtime": 12.1821,
      "eval_samples_per_second": 19.701,
      "eval_steps_per_second": 9.851,
      "step": 9
    },
    {
      "epoch": 0.017543859649122806,
      "grad_norm": 3.516273260116577,
      "learning_rate": 0.0002,
      "loss": 0.8574,
      "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": 2866637124403200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}