checkpoint-5231/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 1000,
  "global_step": 5231,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00019116803670426305,
      "grad_norm": 4.81204080581665,
      "learning_rate": 0.0,
      "loss": 2.4739,
      "step": 1
    },
    {
      "epoch": 0.19116803670426305,
      "grad_norm": 3.8444881439208984,
      "learning_rate": 1.9089171974522293e-05,
      "loss": 1.904,
      "step": 1000
    },
    {
      "epoch": 0.19116803670426305,
      "eval_cosine_accuracy": 0.9566870927810669,
      "eval_loss": 1.0658318996429443,
      "eval_runtime": 24.2832,
      "eval_samples_per_second": 389.817,
      "eval_steps_per_second": 1.524,
      "step": 1000
    },
    {
      "epoch": 0.3823360734085261,
      "grad_norm": 3.6171822547912598,
      "learning_rate": 2.9088720526800256e-05,
      "loss": 1.4035,
      "step": 2000
    },
    {
      "epoch": 0.3823360734085261,
      "eval_cosine_accuracy": 0.9651383757591248,
      "eval_loss": 0.9598055481910706,
      "eval_runtime": 28.2382,
      "eval_samples_per_second": 335.22,
      "eval_steps_per_second": 1.31,
      "step": 2000
    },
    {
      "epoch": 0.5735041101127891,
      "grad_norm": 3.25337290763855,
      "learning_rate": 2.696452595057707e-05,
      "loss": 1.1812,
      "step": 3000
    },
    {
      "epoch": 0.5735041101127891,
      "eval_cosine_accuracy": 0.9709486365318298,
      "eval_loss": 0.9080163836479187,
      "eval_runtime": 23.2786,
      "eval_samples_per_second": 406.639,
      "eval_steps_per_second": 1.589,
      "step": 3000
    },
    {
      "epoch": 0.7646721468170522,
      "grad_norm": 3.5734429359436035,
      "learning_rate": 2.4840331374353893e-05,
      "loss": 1.1325,
      "step": 4000
    },
    {
      "epoch": 0.7646721468170522,
      "eval_cosine_accuracy": 0.9728502035140991,
      "eval_loss": 0.8635017275810242,
      "eval_runtime": 23.4355,
      "eval_samples_per_second": 403.918,
      "eval_steps_per_second": 1.579,
      "step": 4000
    },
    {
      "epoch": 0.9558401835213153,
      "grad_norm": 2.7804830074310303,
      "learning_rate": 2.2718260992706933e-05,
      "loss": 1.6713,
      "step": 5000
    },
    {
      "epoch": 0.9558401835213153,
      "eval_cosine_accuracy": 0.9725332856178284,
      "eval_loss": 0.8610901236534119,
      "eval_runtime": 23.4275,
      "eval_samples_per_second": 404.055,
      "eval_steps_per_second": 1.579,
      "step": 5000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 15693,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}