last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.01002004008016032,
  "eval_steps": 5,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002004008016032064,
      "grad_norm": 0.9981935024261475,
      "learning_rate": 1e-05,
      "loss": 3.6274,
      "step": 1
    },
    {
      "epoch": 0.002004008016032064,
      "eval_loss": 3.705083131790161,
      "eval_runtime": 67.9972,
      "eval_samples_per_second": 24.722,
      "eval_steps_per_second": 12.368,
      "step": 1
    },
    {
      "epoch": 0.004008016032064128,
      "grad_norm": 1.0312002897262573,
      "learning_rate": 2e-05,
      "loss": 3.7344,
      "step": 2
    },
    {
      "epoch": 0.006012024048096192,
      "grad_norm": 1.0591121912002563,
      "learning_rate": 3e-05,
      "loss": 3.6015,
      "step": 3
    },
    {
      "epoch": 0.008016032064128256,
      "grad_norm": 1.0240955352783203,
      "learning_rate": 4e-05,
      "loss": 3.7196,
      "step": 4
    },
    {
      "epoch": 0.01002004008016032,
      "grad_norm": 1.0419278144836426,
      "learning_rate": 5e-05,
      "loss": 3.7358,
      "step": 5
    },
    {
      "epoch": 0.01002004008016032,
      "eval_loss": 3.6875150203704834,
      "eval_runtime": 66.7319,
      "eval_samples_per_second": 25.19,
      "eval_steps_per_second": 12.603,
      "step": 5
    }
  ],
  "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": 1110700688670720.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}