last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.017152658662092625,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0017152658662092624,
      "grad_norm": 10.20623779296875,
      "learning_rate": 2e-05,
      "loss": 10.0968,
      "step": 1
    },
    {
      "epoch": 0.0017152658662092624,
      "eval_loss": 10.568317413330078,
      "eval_runtime": 16.8116,
      "eval_samples_per_second": 14.633,
      "eval_steps_per_second": 7.316,
      "step": 1
    },
    {
      "epoch": 0.003430531732418525,
      "grad_norm": 8.958074569702148,
      "learning_rate": 4e-05,
      "loss": 10.1851,
      "step": 2
    },
    {
      "epoch": 0.005145797598627788,
      "grad_norm": 18.63714599609375,
      "learning_rate": 6e-05,
      "loss": 11.9602,
      "step": 3
    },
    {
      "epoch": 0.005145797598627788,
      "eval_loss": 10.42160701751709,
      "eval_runtime": 16.0368,
      "eval_samples_per_second": 15.34,
      "eval_steps_per_second": 7.67,
      "step": 3
    },
    {
      "epoch": 0.00686106346483705,
      "grad_norm": 15.487518310546875,
      "learning_rate": 8e-05,
      "loss": 10.6799,
      "step": 4
    },
    {
      "epoch": 0.008576329331046312,
      "grad_norm": 12.911466598510742,
      "learning_rate": 0.0001,
      "loss": 9.0754,
      "step": 5
    },
    {
      "epoch": 0.010291595197255575,
      "grad_norm": 14.155526161193848,
      "learning_rate": 0.00012,
      "loss": 8.8563,
      "step": 6
    },
    {
      "epoch": 0.010291595197255575,
      "eval_loss": 7.425621509552002,
      "eval_runtime": 16.0849,
      "eval_samples_per_second": 15.294,
      "eval_steps_per_second": 7.647,
      "step": 6
    },
    {
      "epoch": 0.012006861063464836,
      "grad_norm": 18.84572982788086,
      "learning_rate": 0.00014,
      "loss": 7.9854,
      "step": 7
    },
    {
      "epoch": 0.0137221269296741,
      "grad_norm": 16.23927116394043,
      "learning_rate": 0.00016,
      "loss": 5.275,
      "step": 8
    },
    {
      "epoch": 0.015437392795883362,
      "grad_norm": 21.072099685668945,
      "learning_rate": 0.00018,
      "loss": 4.0998,
      "step": 9
    },
    {
      "epoch": 0.015437392795883362,
      "eval_loss": 3.4490346908569336,
      "eval_runtime": 16.0797,
      "eval_samples_per_second": 15.299,
      "eval_steps_per_second": 7.649,
      "step": 9
    },
    {
      "epoch": 0.017152658662092625,
      "grad_norm": 20.780302047729492,
      "learning_rate": 0.0002,
      "loss": 4.6252,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}