last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.04496065942300487,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0029973772948669914,
      "grad_norm": 1.7252535820007324,
      "learning_rate": 1e-05,
      "loss": 1.1987,
      "step": 1
    },
    {
      "epoch": 0.0029973772948669914,
      "eval_loss": 1.6100515127182007,
      "eval_runtime": 99.8256,
      "eval_samples_per_second": 5.63,
      "eval_steps_per_second": 2.815,
      "step": 1
    },
    {
      "epoch": 0.005994754589733983,
      "grad_norm": 1.6749064922332764,
      "learning_rate": 2e-05,
      "loss": 1.1175,
      "step": 2
    },
    {
      "epoch": 0.008992131884600974,
      "grad_norm": 1.857389211654663,
      "learning_rate": 3e-05,
      "loss": 1.2169,
      "step": 3
    },
    {
      "epoch": 0.011989509179467965,
      "grad_norm": 2.2108614444732666,
      "learning_rate": 4e-05,
      "loss": 1.4463,
      "step": 4
    },
    {
      "epoch": 0.014986886474334956,
      "grad_norm": 2.365875482559204,
      "learning_rate": 5e-05,
      "loss": 1.0604,
      "step": 5
    },
    {
      "epoch": 0.014986886474334956,
      "eval_loss": 1.3768054246902466,
      "eval_runtime": 13.669,
      "eval_samples_per_second": 41.115,
      "eval_steps_per_second": 20.558,
      "step": 5
    },
    {
      "epoch": 0.01798426376920195,
      "grad_norm": 1.8076452016830444,
      "learning_rate": 6e-05,
      "loss": 1.1487,
      "step": 6
    },
    {
      "epoch": 0.02098164106406894,
      "grad_norm": 2.2778992652893066,
      "learning_rate": 7e-05,
      "loss": 1.2715,
      "step": 7
    },
    {
      "epoch": 0.02397901835893593,
      "grad_norm": 2.0399832725524902,
      "learning_rate": 8e-05,
      "loss": 0.8026,
      "step": 8
    },
    {
      "epoch": 0.026976395653802922,
      "grad_norm": 2.653815984725952,
      "learning_rate": 9e-05,
      "loss": 0.7842,
      "step": 9
    },
    {
      "epoch": 0.029973772948669913,
      "grad_norm": 2.3324148654937744,
      "learning_rate": 0.0001,
      "loss": 0.6289,
      "step": 10
    },
    {
      "epoch": 0.029973772948669913,
      "eval_loss": 0.659218966960907,
      "eval_runtime": 13.6822,
      "eval_samples_per_second": 41.075,
      "eval_steps_per_second": 20.538,
      "step": 10
    },
    {
      "epoch": 0.032971150243536904,
      "grad_norm": 2.882046937942505,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.5317,
      "step": 11
    },
    {
      "epoch": 0.0359685275384039,
      "grad_norm": 1.854430913925171,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.4635,
      "step": 12
    },
    {
      "epoch": 0.038965904833270885,
      "grad_norm": 1.4817869663238525,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.2889,
      "step": 13
    },
    {
      "epoch": 0.04196328212813788,
      "grad_norm": 1.171612024307251,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.1823,
      "step": 14
    },
    {
      "epoch": 0.04496065942300487,
      "grad_norm": 1.0793206691741943,
      "learning_rate": 5e-05,
      "loss": 0.1778,
      "step": 15
    },
    {
      "epoch": 0.04496065942300487,
      "eval_loss": 0.1316353678703308,
      "eval_runtime": 13.6689,
      "eval_samples_per_second": 41.115,
      "eval_steps_per_second": 20.558,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1143624595079168e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}