last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0008437928488556059,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.437928488556059e-05,
      "grad_norm": 1.2804934978485107,
      "learning_rate": 2e-05,
      "loss": 2.1467,
      "step": 1
    },
    {
      "epoch": 8.437928488556059e-05,
      "eval_loss": 2.0246927738189697,
      "eval_runtime": 267.5741,
      "eval_samples_per_second": 18.649,
      "eval_steps_per_second": 9.325,
      "step": 1
    },
    {
      "epoch": 0.00016875856977112118,
      "grad_norm": 1.2364039421081543,
      "learning_rate": 4e-05,
      "loss": 2.2813,
      "step": 2
    },
    {
      "epoch": 0.0002531378546566818,
      "grad_norm": 1.0832500457763672,
      "learning_rate": 6e-05,
      "loss": 1.6715,
      "step": 3
    },
    {
      "epoch": 0.0002531378546566818,
      "eval_loss": 2.0154383182525635,
      "eval_runtime": 266.0975,
      "eval_samples_per_second": 18.753,
      "eval_steps_per_second": 9.376,
      "step": 3
    },
    {
      "epoch": 0.00033751713954224236,
      "grad_norm": 1.1004549264907837,
      "learning_rate": 8e-05,
      "loss": 1.4839,
      "step": 4
    },
    {
      "epoch": 0.00042189642442780296,
      "grad_norm": 1.5068570375442505,
      "learning_rate": 0.0001,
      "loss": 1.9365,
      "step": 5
    },
    {
      "epoch": 0.0005062757093133636,
      "grad_norm": 1.5566529035568237,
      "learning_rate": 0.00012,
      "loss": 1.7985,
      "step": 6
    },
    {
      "epoch": 0.0005062757093133636,
      "eval_loss": 1.8656870126724243,
      "eval_runtime": 266.2578,
      "eval_samples_per_second": 18.741,
      "eval_steps_per_second": 9.371,
      "step": 6
    },
    {
      "epoch": 0.0005906549941989242,
      "grad_norm": 1.7015923261642456,
      "learning_rate": 0.00014,
      "loss": 1.4552,
      "step": 7
    },
    {
      "epoch": 0.0006750342790844847,
      "grad_norm": 2.224033832550049,
      "learning_rate": 0.00016,
      "loss": 1.4939,
      "step": 8
    },
    {
      "epoch": 0.0007594135639700454,
      "grad_norm": 1.8080663681030273,
      "learning_rate": 0.00018,
      "loss": 1.2411,
      "step": 9
    },
    {
      "epoch": 0.0007594135639700454,
      "eval_loss": 1.3588569164276123,
      "eval_runtime": 266.3078,
      "eval_samples_per_second": 18.738,
      "eval_steps_per_second": 9.369,
      "step": 9
    },
    {
      "epoch": 0.0008437928488556059,
      "grad_norm": 1.9288817644119263,
      "learning_rate": 0.0002,
      "loss": 1.1243,
      "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": 3710332226764800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}