last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00023609127288609777,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.3609127288609778e-05,
      "grad_norm": 3.169210433959961,
      "learning_rate": 2e-05,
      "loss": 1.8968,
      "step": 1
    },
    {
      "epoch": 2.3609127288609778e-05,
      "eval_loss": 2.400615930557251,
      "eval_runtime": 524.7981,
      "eval_samples_per_second": 16.993,
      "eval_steps_per_second": 16.993,
      "step": 1
    },
    {
      "epoch": 4.7218254577219556e-05,
      "grad_norm": 3.1677653789520264,
      "learning_rate": 4e-05,
      "loss": 1.7531,
      "step": 2
    },
    {
      "epoch": 7.082738186582933e-05,
      "grad_norm": 4.297532558441162,
      "learning_rate": 6e-05,
      "loss": 2.8112,
      "step": 3
    },
    {
      "epoch": 7.082738186582933e-05,
      "eval_loss": 2.3562111854553223,
      "eval_runtime": 524.4667,
      "eval_samples_per_second": 17.004,
      "eval_steps_per_second": 17.004,
      "step": 3
    },
    {
      "epoch": 9.443650915443911e-05,
      "grad_norm": 3.5680267810821533,
      "learning_rate": 8e-05,
      "loss": 2.2052,
      "step": 4
    },
    {
      "epoch": 0.00011804563644304888,
      "grad_norm": 3.6056132316589355,
      "learning_rate": 0.0001,
      "loss": 1.8426,
      "step": 5
    },
    {
      "epoch": 0.00014165476373165865,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 2.1509,
      "step": 6
    },
    {
      "epoch": 0.00014165476373165865,
      "eval_loss": 2.2560741901397705,
      "eval_runtime": 524.6957,
      "eval_samples_per_second": 16.997,
      "eval_steps_per_second": 16.997,
      "step": 6
    },
    {
      "epoch": 0.00016526389102026844,
      "grad_norm": 4.292576789855957,
      "learning_rate": 0.00012,
      "loss": 2.4561,
      "step": 7
    },
    {
      "epoch": 0.00018887301830887822,
      "grad_norm": 5.075192928314209,
      "learning_rate": 0.00014,
      "loss": 2.4012,
      "step": 8
    },
    {
      "epoch": 0.00021248214559748798,
      "grad_norm": 4.3318610191345215,
      "learning_rate": 0.00016,
      "loss": 1.9781,
      "step": 9
    },
    {
      "epoch": 0.00021248214559748798,
      "eval_loss": 2.1713130474090576,
      "eval_runtime": 524.5919,
      "eval_samples_per_second": 17.0,
      "eval_steps_per_second": 17.0,
      "step": 9
    },
    {
      "epoch": 0.00023609127288609777,
      "grad_norm": 4.921598434448242,
      "learning_rate": 0.00018,
      "loss": 2.7927,
      "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": 360477150412800.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}