checkpoint-1000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.2753128555176336,
  "eval_steps": 1000,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11376564277588168,
      "grad_norm": 0.24430778622627258,
      "learning_rate": 9.99848701000714e-05,
      "loss": 1.1872,
      "step": 50
    },
    {
      "epoch": 0.22753128555176336,
      "grad_norm": 0.2623399794101715,
      "learning_rate": 9.945628442036389e-05,
      "loss": 1.0129,
      "step": 100
    },
    {
      "epoch": 0.3412969283276451,
      "grad_norm": 0.27533629536628723,
      "learning_rate": 9.818033542127878e-05,
      "loss": 0.9522,
      "step": 150
    },
    {
      "epoch": 0.4550625711035267,
      "grad_norm": 0.3079286813735962,
      "learning_rate": 9.61763047267894e-05,
      "loss": 0.9455,
      "step": 200
    },
    {
      "epoch": 0.5688282138794084,
      "grad_norm": 0.30581459403038025,
      "learning_rate": 9.347447643626802e-05,
      "loss": 0.9141,
      "step": 250
    },
    {
      "epoch": 0.6825938566552902,
      "grad_norm": 0.3370199501514435,
      "learning_rate": 9.011567948345395e-05,
      "loss": 0.9212,
      "step": 300
    },
    {
      "epoch": 0.7963594994311718,
      "grad_norm": 0.3328758776187897,
      "learning_rate": 8.615067064615998e-05,
      "loss": 0.886,
      "step": 350
    },
    {
      "epoch": 0.9101251422070534,
      "grad_norm": 0.3751530647277832,
      "learning_rate": 8.16393675304241e-05,
      "loss": 0.8991,
      "step": 400
    },
    {
      "epoch": 1.023890784982935,
      "grad_norm": 0.33782124519348145,
      "learning_rate": 7.66499431199391e-05,
      "loss": 0.8787,
      "step": 450
    },
    {
      "epoch": 1.1376564277588168,
      "grad_norm": 0.3481363356113434,
      "learning_rate": 7.125779557356193e-05,
      "loss": 0.8458,
      "step": 500
    },
    {
      "epoch": 1.2514220705346986,
      "grad_norm": 0.41307157278060913,
      "learning_rate": 6.554440883890547e-05,
      "loss": 0.8482,
      "step": 550
    },
    {
      "epoch": 1.36518771331058,
      "grad_norm": 0.4020177721977234,
      "learning_rate": 5.959612129995763e-05,
      "loss": 0.8467,
      "step": 600
    },
    {
      "epoch": 1.4789533560864618,
      "grad_norm": 0.3888244926929474,
      "learning_rate": 5.3502821066426265e-05,
      "loss": 0.8151,
      "step": 650
    },
    {
      "epoch": 1.5927189988623436,
      "grad_norm": 0.391117662191391,
      "learning_rate": 4.7356587621068075e-05,
      "loss": 0.822,
      "step": 700
    },
    {
      "epoch": 1.7064846416382253,
      "grad_norm": 0.4161781668663025,
      "learning_rate": 4.125030035187653e-05,
      "loss": 0.7955,
      "step": 750
    },
    {
      "epoch": 1.820250284414107,
      "grad_norm": 0.4105525314807892,
      "learning_rate": 3.527623499642595e-05,
      "loss": 0.8151,
      "step": 800
    },
    {
      "epoch": 1.9340159271899886,
      "grad_norm": 0.41017189621925354,
      "learning_rate": 2.952466920833622e-05,
      "loss": 0.824,
      "step": 850
    },
    {
      "epoch": 2.04778156996587,
      "grad_norm": 0.41780519485473633,
      "learning_rate": 2.408251831797206e-05,
      "loss": 0.7866,
      "step": 900
    },
    {
      "epoch": 2.161547212741752,
      "grad_norm": 0.4392172396183014,
      "learning_rate": 1.903202190320803e-05,
      "loss": 0.7672,
      "step": 950
    },
    {
      "epoch": 2.2753128555176336,
      "grad_norm": 0.45009684562683105,
      "learning_rate": 1.4449501018269717e-05,
      "loss": 0.763,
      "step": 1000
    },
    {
      "epoch": 2.2753128555176336,
      "eval_loss": 0.8412221074104309,
      "eval_runtime": 215.6811,
      "eval_samples_per_second": 0.923,
      "eval_steps_per_second": 0.923,
      "step": 1000
    }
  ],
  "logging_steps": 50,
  "max_steps": 1317,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.7075818133428634e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}