last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0008215914225855482,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.215914225855481e-05,
      "grad_norm": 1.0494390726089478,
      "learning_rate": 2e-05,
      "loss": 0.7067,
      "step": 1
    },
    {
      "epoch": 8.215914225855481e-05,
      "eval_loss": 1.0838401317596436,
      "eval_runtime": 597.7705,
      "eval_samples_per_second": 4.288,
      "eval_steps_per_second": 4.288,
      "step": 1
    },
    {
      "epoch": 0.00016431828451710963,
      "grad_norm": 1.048064112663269,
      "learning_rate": 4e-05,
      "loss": 0.9619,
      "step": 2
    },
    {
      "epoch": 0.00024647742677566444,
      "grad_norm": 1.6032434701919556,
      "learning_rate": 6e-05,
      "loss": 0.9577,
      "step": 3
    },
    {
      "epoch": 0.00024647742677566444,
      "eval_loss": 1.0675935745239258,
      "eval_runtime": 598.3776,
      "eval_samples_per_second": 4.283,
      "eval_steps_per_second": 4.283,
      "step": 3
    },
    {
      "epoch": 0.00032863656903421926,
      "grad_norm": 1.337504267692566,
      "learning_rate": 8e-05,
      "loss": 0.7016,
      "step": 4
    },
    {
      "epoch": 0.0004107957112927741,
      "grad_norm": 1.7052624225616455,
      "learning_rate": 0.0001,
      "loss": 0.9557,
      "step": 5
    },
    {
      "epoch": 0.0004929548535513289,
      "grad_norm": 1.0046615600585938,
      "learning_rate": 0.00012,
      "loss": 0.855,
      "step": 6
    },
    {
      "epoch": 0.0004929548535513289,
      "eval_loss": 1.0117896795272827,
      "eval_runtime": 598.3813,
      "eval_samples_per_second": 4.283,
      "eval_steps_per_second": 4.283,
      "step": 6
    },
    {
      "epoch": 0.0005751139958098837,
      "grad_norm": 1.472562313079834,
      "learning_rate": 0.00014,
      "loss": 0.8744,
      "step": 7
    },
    {
      "epoch": 0.0006572731380684385,
      "grad_norm": 0.7952223420143127,
      "learning_rate": 0.00016,
      "loss": 0.4751,
      "step": 8
    },
    {
      "epoch": 0.0007394322803269934,
      "grad_norm": 1.635096788406372,
      "learning_rate": 0.00018,
      "loss": 0.863,
      "step": 9
    },
    {
      "epoch": 0.0007394322803269934,
      "eval_loss": 0.9735259413719177,
      "eval_runtime": 599.4848,
      "eval_samples_per_second": 4.275,
      "eval_steps_per_second": 4.275,
      "step": 9
    },
    {
      "epoch": 0.0008215914225855482,
      "grad_norm": 1.9233784675598145,
      "learning_rate": 0.0002,
      "loss": 0.6987,
      "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": 7668943615426560.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}