last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.011128284389489953,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0012364760432766616,
      "grad_norm": 13.879922866821289,
      "learning_rate": 2e-05,
      "loss": 6.7123,
      "step": 1
    },
    {
      "epoch": 0.0012364760432766616,
      "eval_loss": 5.792558193206787,
      "eval_runtime": 72.4872,
      "eval_samples_per_second": 2.359,
      "eval_steps_per_second": 2.359,
      "step": 1
    },
    {
      "epoch": 0.002472952086553323,
      "grad_norm": 21.114770889282227,
      "learning_rate": 4e-05,
      "loss": 5.7859,
      "step": 2
    },
    {
      "epoch": 0.0037094281298299847,
      "grad_norm": 19.899456024169922,
      "learning_rate": 6e-05,
      "loss": 5.8473,
      "step": 3
    },
    {
      "epoch": 0.0037094281298299847,
      "eval_loss": 5.629321575164795,
      "eval_runtime": 73.572,
      "eval_samples_per_second": 2.324,
      "eval_steps_per_second": 2.324,
      "step": 3
    },
    {
      "epoch": 0.004945904173106646,
      "grad_norm": 29.271102905273438,
      "learning_rate": 8e-05,
      "loss": 5.6493,
      "step": 4
    },
    {
      "epoch": 0.0061823802163833074,
      "grad_norm": 34.93131637573242,
      "learning_rate": 0.0001,
      "loss": 3.9794,
      "step": 5
    },
    {
      "epoch": 0.0074188562596599695,
      "grad_norm": 30.170530319213867,
      "learning_rate": 0.00012,
      "loss": 4.9902,
      "step": 6
    },
    {
      "epoch": 0.0074188562596599695,
      "eval_loss": 3.892670154571533,
      "eval_runtime": 73.758,
      "eval_samples_per_second": 2.318,
      "eval_steps_per_second": 2.318,
      "step": 6
    },
    {
      "epoch": 0.00865533230293663,
      "grad_norm": 12.877334594726562,
      "learning_rate": 0.00014,
      "loss": 3.5622,
      "step": 7
    },
    {
      "epoch": 0.009891808346213293,
      "grad_norm": 24.434354782104492,
      "learning_rate": 0.00016,
      "loss": 3.8512,
      "step": 8
    },
    {
      "epoch": 0.011128284389489953,
      "grad_norm": 16.24271583557129,
      "learning_rate": 0.00018,
      "loss": 3.0013,
      "step": 9
    },
    {
      "epoch": 0.011128284389489953,
      "eval_loss": 2.780280113220215,
      "eval_runtime": 73.7018,
      "eval_samples_per_second": 2.32,
      "eval_steps_per_second": 2.32,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6676985552044032.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}