last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.015625,
  "eval_steps": 8,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005208333333333333,
      "eval_loss": 5.936483383178711,
      "eval_runtime": 75.5811,
      "eval_samples_per_second": 10.704,
      "eval_steps_per_second": 5.358,
      "step": 1
    },
    {
      "epoch": 0.0015625,
      "grad_norm": 9.487432479858398,
      "learning_rate": 6e-05,
      "loss": 5.6909,
      "step": 3
    },
    {
      "epoch": 0.003125,
      "grad_norm": 7.271123886108398,
      "learning_rate": 0.00012,
      "loss": 4.7782,
      "step": 6
    },
    {
      "epoch": 0.004166666666666667,
      "eval_loss": 3.7537190914154053,
      "eval_runtime": 76.702,
      "eval_samples_per_second": 10.547,
      "eval_steps_per_second": 5.28,
      "step": 8
    },
    {
      "epoch": 0.0046875,
      "grad_norm": 9.870651245117188,
      "learning_rate": 0.00018,
      "loss": 4.0272,
      "step": 9
    },
    {
      "epoch": 0.00625,
      "grad_norm": 3.0841119289398193,
      "learning_rate": 0.00019510565162951537,
      "loss": 2.7237,
      "step": 12
    },
    {
      "epoch": 0.0078125,
      "grad_norm": 3.261676549911499,
      "learning_rate": 0.00017071067811865476,
      "loss": 2.7135,
      "step": 15
    },
    {
      "epoch": 0.008333333333333333,
      "eval_loss": 2.524137020111084,
      "eval_runtime": 76.8339,
      "eval_samples_per_second": 10.529,
      "eval_steps_per_second": 5.271,
      "step": 16
    },
    {
      "epoch": 0.009375,
      "grad_norm": 3.2629971504211426,
      "learning_rate": 0.00013090169943749476,
      "loss": 2.4572,
      "step": 18
    },
    {
      "epoch": 0.0109375,
      "grad_norm": 2.834351062774658,
      "learning_rate": 8.435655349597689e-05,
      "loss": 2.4698,
      "step": 21
    },
    {
      "epoch": 0.0125,
      "grad_norm": 2.5578553676605225,
      "learning_rate": 4.12214747707527e-05,
      "loss": 2.3436,
      "step": 24
    },
    {
      "epoch": 0.0125,
      "eval_loss": 2.4004898071289062,
      "eval_runtime": 76.8553,
      "eval_samples_per_second": 10.526,
      "eval_steps_per_second": 5.27,
      "step": 24
    },
    {
      "epoch": 0.0140625,
      "grad_norm": 2.488203763961792,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 2.496,
      "step": 27
    },
    {
      "epoch": 0.015625,
      "grad_norm": 2.202345132827759,
      "learning_rate": 0.0,
      "loss": 2.3414,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.150341542313984e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}