last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.002975304968759298,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002975304968759298,
      "grad_norm": 4.083354473114014,
      "learning_rate": 2e-05,
      "loss": 6.2338,
      "step": 1
    },
    {
      "epoch": 0.0002975304968759298,
      "eval_loss": 3.9553987979888916,
      "eval_runtime": 170.2808,
      "eval_samples_per_second": 8.316,
      "eval_steps_per_second": 4.158,
      "step": 1
    },
    {
      "epoch": 0.0005950609937518596,
      "grad_norm": 2.295527219772339,
      "learning_rate": 4e-05,
      "loss": 3.6333,
      "step": 2
    },
    {
      "epoch": 0.0008925914906277894,
      "grad_norm": 2.81231689453125,
      "learning_rate": 6e-05,
      "loss": 4.7203,
      "step": 3
    },
    {
      "epoch": 0.0008925914906277894,
      "eval_loss": 3.942405939102173,
      "eval_runtime": 172.0013,
      "eval_samples_per_second": 8.232,
      "eval_steps_per_second": 4.116,
      "step": 3
    },
    {
      "epoch": 0.0011901219875037191,
      "grad_norm": 2.846064805984497,
      "learning_rate": 8e-05,
      "loss": 4.558,
      "step": 4
    },
    {
      "epoch": 0.001487652484379649,
      "grad_norm": 1.77438223361969,
      "learning_rate": 0.0001,
      "loss": 3.2527,
      "step": 5
    },
    {
      "epoch": 0.0017851829812555787,
      "grad_norm": 2.6561055183410645,
      "learning_rate": 0.00012,
      "loss": 4.8197,
      "step": 6
    },
    {
      "epoch": 0.0017851829812555787,
      "eval_loss": 3.7654619216918945,
      "eval_runtime": 171.8954,
      "eval_samples_per_second": 8.238,
      "eval_steps_per_second": 4.119,
      "step": 6
    },
    {
      "epoch": 0.0020827134781315083,
      "grad_norm": 1.1815612316131592,
      "learning_rate": 0.00014,
      "loss": 2.215,
      "step": 7
    },
    {
      "epoch": 0.0023802439750074383,
      "grad_norm": 2.3964359760284424,
      "learning_rate": 0.00016,
      "loss": 3.0826,
      "step": 8
    },
    {
      "epoch": 0.002677774471883368,
      "grad_norm": 4.077512264251709,
      "learning_rate": 0.00018,
      "loss": 3.6421,
      "step": 9
    },
    {
      "epoch": 0.002677774471883368,
      "eval_loss": 3.4991986751556396,
      "eval_runtime": 172.0593,
      "eval_samples_per_second": 8.23,
      "eval_steps_per_second": 4.115,
      "step": 9
    },
    {
      "epoch": 0.002975304968759298,
      "grad_norm": 2.2527685165405273,
      "learning_rate": 0.0002,
      "loss": 2.4711,
      "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": 2882232561696768.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}