last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.05161290322580645,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005161290322580645,
      "grad_norm": 2.1501665115356445,
      "learning_rate": 2e-05,
      "loss": 2.8344,
      "step": 1
    },
    {
      "epoch": 0.005161290322580645,
      "eval_loss": 3.26839542388916,
      "eval_runtime": 9.8046,
      "eval_samples_per_second": 16.727,
      "eval_steps_per_second": 4.182,
      "step": 1
    },
    {
      "epoch": 0.01032258064516129,
      "grad_norm": 2.0289418697357178,
      "learning_rate": 4e-05,
      "loss": 3.3252,
      "step": 2
    },
    {
      "epoch": 0.015483870967741935,
      "grad_norm": 2.006923198699951,
      "learning_rate": 6e-05,
      "loss": 2.9502,
      "step": 3
    },
    {
      "epoch": 0.015483870967741935,
      "eval_loss": 3.10579776763916,
      "eval_runtime": 9.2067,
      "eval_samples_per_second": 17.813,
      "eval_steps_per_second": 4.453,
      "step": 3
    },
    {
      "epoch": 0.02064516129032258,
      "grad_norm": 2.4136576652526855,
      "learning_rate": 8e-05,
      "loss": 2.7058,
      "step": 4
    },
    {
      "epoch": 0.025806451612903226,
      "grad_norm": 2.2730557918548584,
      "learning_rate": 0.0001,
      "loss": 2.881,
      "step": 5
    },
    {
      "epoch": 0.03096774193548387,
      "grad_norm": 2.3009700775146484,
      "learning_rate": 0.00012,
      "loss": 2.5958,
      "step": 6
    },
    {
      "epoch": 0.03096774193548387,
      "eval_loss": 2.5263447761535645,
      "eval_runtime": 9.2214,
      "eval_samples_per_second": 17.785,
      "eval_steps_per_second": 4.446,
      "step": 6
    },
    {
      "epoch": 0.03612903225806452,
      "grad_norm": 3.047846794128418,
      "learning_rate": 0.00014,
      "loss": 2.3486,
      "step": 7
    },
    {
      "epoch": 0.04129032258064516,
      "grad_norm": 3.2430038452148438,
      "learning_rate": 0.00016,
      "loss": 2.1589,
      "step": 8
    },
    {
      "epoch": 0.04645161290322581,
      "grad_norm": 2.8587591648101807,
      "learning_rate": 0.00018,
      "loss": 2.1334,
      "step": 9
    },
    {
      "epoch": 0.04645161290322581,
      "eval_loss": 1.8293477296829224,
      "eval_runtime": 9.2285,
      "eval_samples_per_second": 17.771,
      "eval_steps_per_second": 4.443,
      "step": 9
    },
    {
      "epoch": 0.05161290322580645,
      "grad_norm": 1.926283359527588,
      "learning_rate": 0.0002,
      "loss": 1.6968,
      "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": 2.796154706198528e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}