last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.042417815482502653,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0042417815482502655,
      "grad_norm": 30.247900009155273,
      "learning_rate": 2e-05,
      "loss": 12.4376,
      "step": 1
    },
    {
      "epoch": 0.0042417815482502655,
      "eval_loss": 11.358438491821289,
      "eval_runtime": 2.9321,
      "eval_samples_per_second": 34.105,
      "eval_steps_per_second": 17.053,
      "step": 1
    },
    {
      "epoch": 0.008483563096500531,
      "grad_norm": 30.898862838745117,
      "learning_rate": 4e-05,
      "loss": 11.5013,
      "step": 2
    },
    {
      "epoch": 0.012725344644750796,
      "grad_norm": 30.758533477783203,
      "learning_rate": 6e-05,
      "loss": 11.1251,
      "step": 3
    },
    {
      "epoch": 0.012725344644750796,
      "eval_loss": 10.850805282592773,
      "eval_runtime": 2.9205,
      "eval_samples_per_second": 34.24,
      "eval_steps_per_second": 17.12,
      "step": 3
    },
    {
      "epoch": 0.016967126193001062,
      "grad_norm": 27.131473541259766,
      "learning_rate": 8e-05,
      "loss": 10.742,
      "step": 4
    },
    {
      "epoch": 0.021208907741251327,
      "grad_norm": 33.091026306152344,
      "learning_rate": 0.0001,
      "loss": 10.1742,
      "step": 5
    },
    {
      "epoch": 0.02545068928950159,
      "grad_norm": 31.555877685546875,
      "learning_rate": 0.00012,
      "loss": 8.5343,
      "step": 6
    },
    {
      "epoch": 0.02545068928950159,
      "eval_loss": 5.258501052856445,
      "eval_runtime": 3.0006,
      "eval_samples_per_second": 33.327,
      "eval_steps_per_second": 16.663,
      "step": 6
    },
    {
      "epoch": 0.029692470837751856,
      "grad_norm": 22.82707977294922,
      "learning_rate": 0.00014,
      "loss": 5.0809,
      "step": 7
    },
    {
      "epoch": 0.033934252386002124,
      "grad_norm": 16.72810935974121,
      "learning_rate": 0.00016,
      "loss": 2.8931,
      "step": 8
    },
    {
      "epoch": 0.03817603393425239,
      "grad_norm": 13.084162712097168,
      "learning_rate": 0.00018,
      "loss": 2.3121,
      "step": 9
    },
    {
      "epoch": 0.03817603393425239,
      "eval_loss": 1.4584784507751465,
      "eval_runtime": 3.0188,
      "eval_samples_per_second": 33.125,
      "eval_steps_per_second": 16.563,
      "step": 9
    },
    {
      "epoch": 0.042417815482502653,
      "grad_norm": 9.421570777893066,
      "learning_rate": 0.0002,
      "loss": 1.1165,
      "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": 500097330708480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}