last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0032526936369180726,
  "eval_steps": 8,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00016263468184590365,
      "eval_loss": 0.6653284430503845,
      "eval_runtime": 324.3776,
      "eval_samples_per_second": 7.981,
      "eval_steps_per_second": 3.992,
      "step": 1
    },
    {
      "epoch": 0.0004879040455377109,
      "grad_norm": 0.656542956829071,
      "learning_rate": 6e-05,
      "loss": 2.8969,
      "step": 3
    },
    {
      "epoch": 0.0009758080910754218,
      "grad_norm": 0.7154726982116699,
      "learning_rate": 0.00012,
      "loss": 2.6609,
      "step": 6
    },
    {
      "epoch": 0.0013010774547672292,
      "eval_loss": 0.6452193260192871,
      "eval_runtime": 324.1104,
      "eval_samples_per_second": 7.988,
      "eval_steps_per_second": 3.996,
      "step": 8
    },
    {
      "epoch": 0.0014637121366131327,
      "grad_norm": 0.733124315738678,
      "learning_rate": 0.00018,
      "loss": 2.5338,
      "step": 9
    },
    {
      "epoch": 0.0019516161821508437,
      "grad_norm": 1.2633097171783447,
      "learning_rate": 0.00019510565162951537,
      "loss": 2.7224,
      "step": 12
    },
    {
      "epoch": 0.002439520227688555,
      "grad_norm": 0.614086389541626,
      "learning_rate": 0.00017071067811865476,
      "loss": 2.7872,
      "step": 15
    },
    {
      "epoch": 0.0026021549095344584,
      "eval_loss": 0.6183476448059082,
      "eval_runtime": 323.9576,
      "eval_samples_per_second": 7.992,
      "eval_steps_per_second": 3.997,
      "step": 16
    },
    {
      "epoch": 0.0029274242732262655,
      "grad_norm": 0.567788302898407,
      "learning_rate": 0.00013090169943749476,
      "loss": 2.1966,
      "step": 18
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.856740263985152e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}