trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9752650176678443,
  "eval_steps": 500,
  "global_step": 105,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2826855123674912,
      "grad_norm": 4.498367901417786,
      "learning_rate": 5e-06,
      "loss": 1.1885,
      "step": 10
    },
    {
      "epoch": 0.5653710247349824,
      "grad_norm": 2.901944412811591,
      "learning_rate": 5e-06,
      "loss": 1.0832,
      "step": 20
    },
    {
      "epoch": 0.8480565371024735,
      "grad_norm": 5.838345247279183,
      "learning_rate": 5e-06,
      "loss": 1.0481,
      "step": 30
    },
    {
      "epoch": 0.9893992932862191,
      "eval_loss": 1.0256564617156982,
      "eval_runtime": 24.17,
      "eval_samples_per_second": 39.346,
      "eval_steps_per_second": 0.621,
      "step": 35
    },
    {
      "epoch": 1.1342756183745584,
      "grad_norm": 5.345092823125488,
      "learning_rate": 5e-06,
      "loss": 1.092,
      "step": 40
    },
    {
      "epoch": 1.4169611307420495,
      "grad_norm": 1.536337643464911,
      "learning_rate": 5e-06,
      "loss": 0.9837,
      "step": 50
    },
    {
      "epoch": 1.6996466431095407,
      "grad_norm": 1.2189970943931634,
      "learning_rate": 5e-06,
      "loss": 0.9671,
      "step": 60
    },
    {
      "epoch": 1.9823321554770317,
      "grad_norm": 1.199061287119998,
      "learning_rate": 5e-06,
      "loss": 0.9613,
      "step": 70
    },
    {
      "epoch": 1.9823321554770317,
      "eval_loss": 0.9848190546035767,
      "eval_runtime": 24.1082,
      "eval_samples_per_second": 39.447,
      "eval_steps_per_second": 0.622,
      "step": 70
    },
    {
      "epoch": 2.2685512367491167,
      "grad_norm": 2.2674396976461337,
      "learning_rate": 5e-06,
      "loss": 0.9866,
      "step": 80
    },
    {
      "epoch": 2.5512367491166077,
      "grad_norm": 0.8991668226108301,
      "learning_rate": 5e-06,
      "loss": 0.9045,
      "step": 90
    },
    {
      "epoch": 2.833922261484099,
      "grad_norm": 0.7687134493215432,
      "learning_rate": 5e-06,
      "loss": 0.8991,
      "step": 100
    },
    {
      "epoch": 2.9752650176678443,
      "eval_loss": 0.9663988947868347,
      "eval_runtime": 23.2532,
      "eval_samples_per_second": 40.898,
      "eval_steps_per_second": 0.645,
      "step": 105
    },
    {
      "epoch": 2.9752650176678443,
      "step": 105,
      "total_flos": 175669531115520.0,
      "train_loss": 1.0058174587431408,
      "train_runtime": 3745.1735,
      "train_samples_per_second": 14.465,
      "train_steps_per_second": 0.028
    }
  ],
  "logging_steps": 10,
  "max_steps": 105,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 175669531115520.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}