last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.006278449223041909,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006278449223041908,
      "grad_norm": 1.6742286682128906,
      "learning_rate": 2e-05,
      "loss": 7.3124,
      "step": 1
    },
    {
      "epoch": 0.0006278449223041908,
      "eval_loss": 2.1191673278808594,
      "eval_runtime": 35.9183,
      "eval_samples_per_second": 18.681,
      "eval_steps_per_second": 9.355,
      "step": 1
    },
    {
      "epoch": 0.0012556898446083817,
      "grad_norm": 2.8826026916503906,
      "learning_rate": 4e-05,
      "loss": 10.6008,
      "step": 2
    },
    {
      "epoch": 0.0018835347669125726,
      "grad_norm": 2.1440272331237793,
      "learning_rate": 6e-05,
      "loss": 7.9409,
      "step": 3
    },
    {
      "epoch": 0.0018835347669125726,
      "eval_loss": 2.1142067909240723,
      "eval_runtime": 35.9015,
      "eval_samples_per_second": 18.69,
      "eval_steps_per_second": 9.359,
      "step": 3
    },
    {
      "epoch": 0.0025113796892167633,
      "grad_norm": 1.8994042873382568,
      "learning_rate": 8e-05,
      "loss": 6.5246,
      "step": 4
    },
    {
      "epoch": 0.0031392246115209545,
      "grad_norm": 2.968654155731201,
      "learning_rate": 0.0001,
      "loss": 10.2607,
      "step": 5
    },
    {
      "epoch": 0.003767069533825145,
      "grad_norm": 3.711641311645508,
      "learning_rate": 0.00012,
      "loss": 8.121,
      "step": 6
    },
    {
      "epoch": 0.003767069533825145,
      "eval_loss": 2.017463445663452,
      "eval_runtime": 35.9333,
      "eval_samples_per_second": 18.674,
      "eval_steps_per_second": 9.351,
      "step": 6
    },
    {
      "epoch": 0.004394914456129336,
      "grad_norm": 4.26845645904541,
      "learning_rate": 0.00014,
      "loss": 10.8731,
      "step": 7
    },
    {
      "epoch": 0.005022759378433527,
      "grad_norm": 4.642733573913574,
      "learning_rate": 0.00016,
      "loss": 8.4455,
      "step": 8
    },
    {
      "epoch": 0.005650604300737718,
      "grad_norm": 2.958479642868042,
      "learning_rate": 0.00018,
      "loss": 5.2765,
      "step": 9
    },
    {
      "epoch": 0.005650604300737718,
      "eval_loss": 1.4980762004852295,
      "eval_runtime": 35.915,
      "eval_samples_per_second": 18.683,
      "eval_steps_per_second": 9.355,
      "step": 9
    },
    {
      "epoch": 0.006278449223041909,
      "grad_norm": 3.713488817214966,
      "learning_rate": 0.0002,
      "loss": 5.4786,
      "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": 3166224305356800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}