checkpoint-99/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 10,
  "global_step": 99,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 0.17900869250297546,
      "learning_rate": 5e-05,
      "loss": 1.7452,
      "step": 10
    },
    {
      "epoch": 0.30303030303030304,
      "eval_loss": 1.7232011556625366,
      "eval_runtime": 49.4935,
      "eval_samples_per_second": 2.283,
      "eval_steps_per_second": 1.152,
      "step": 10
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 0.2701665163040161,
      "learning_rate": 5e-05,
      "loss": 1.6822,
      "step": 20
    },
    {
      "epoch": 0.6060606060606061,
      "eval_loss": 1.6173911094665527,
      "eval_runtime": 58.7659,
      "eval_samples_per_second": 1.923,
      "eval_steps_per_second": 0.97,
      "step": 20
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.3478323519229889,
      "learning_rate": 5e-05,
      "loss": 1.5466,
      "step": 30
    },
    {
      "epoch": 0.9090909090909091,
      "eval_loss": 1.4601249694824219,
      "eval_runtime": 49.5747,
      "eval_samples_per_second": 2.279,
      "eval_steps_per_second": 1.15,
      "step": 30
    },
    {
      "epoch": 1.2121212121212122,
      "grad_norm": 0.5212990045547485,
      "learning_rate": 5e-05,
      "loss": 1.3684,
      "step": 40
    },
    {
      "epoch": 1.2121212121212122,
      "eval_loss": 1.2638370990753174,
      "eval_runtime": 49.4777,
      "eval_samples_per_second": 2.284,
      "eval_steps_per_second": 1.152,
      "step": 40
    },
    {
      "epoch": 1.5151515151515151,
      "grad_norm": 0.8404479026794434,
      "learning_rate": 5e-05,
      "loss": 1.1522,
      "step": 50
    },
    {
      "epoch": 1.5151515151515151,
      "eval_loss": 1.0334552526474,
      "eval_runtime": 50.0219,
      "eval_samples_per_second": 2.259,
      "eval_steps_per_second": 1.14,
      "step": 50
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 1.5856136083602905,
      "learning_rate": 5e-05,
      "loss": 0.9142,
      "step": 60
    },
    {
      "epoch": 1.8181818181818183,
      "eval_loss": 0.784606397151947,
      "eval_runtime": 65.2695,
      "eval_samples_per_second": 1.731,
      "eval_steps_per_second": 0.873,
      "step": 60
    },
    {
      "epoch": 2.121212121212121,
      "grad_norm": 1.629359483718872,
      "learning_rate": 5e-05,
      "loss": 0.6577,
      "step": 70
    },
    {
      "epoch": 2.121212121212121,
      "eval_loss": 0.5981450080871582,
      "eval_runtime": 49.0711,
      "eval_samples_per_second": 2.303,
      "eval_steps_per_second": 1.162,
      "step": 70
    },
    {
      "epoch": 2.4242424242424243,
      "grad_norm": 1.0707775354385376,
      "learning_rate": 5e-05,
      "loss": 0.5194,
      "step": 80
    },
    {
      "epoch": 2.4242424242424243,
      "eval_loss": 0.4855763018131256,
      "eval_runtime": 49.0294,
      "eval_samples_per_second": 2.305,
      "eval_steps_per_second": 1.163,
      "step": 80
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.605841338634491,
      "learning_rate": 5e-05,
      "loss": 0.4489,
      "step": 90
    },
    {
      "epoch": 2.7272727272727275,
      "eval_loss": 0.4201233386993408,
      "eval_runtime": 49.0709,
      "eval_samples_per_second": 2.303,
      "eval_steps_per_second": 1.162,
      "step": 90
    }
  ],
  "logging_steps": 10,
  "max_steps": 99,
  "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": 7.46511553880064e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}