checkpoint-3984/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 3984,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7530120481927711,
      "grad_norm": 0.2647170126438141,
      "learning_rate": 0.0004811746987951807,
      "loss": 0.3311,
      "step": 500
    },
    {
      "epoch": 1.5060240963855422,
      "grad_norm": 0.22880347073078156,
      "learning_rate": 0.00046234939759036143,
      "loss": 0.0907,
      "step": 1000
    },
    {
      "epoch": 2.2590361445783134,
      "grad_norm": 0.1677163541316986,
      "learning_rate": 0.00044352409638554217,
      "loss": 0.0568,
      "step": 1500
    },
    {
      "epoch": 3.0120481927710845,
      "grad_norm": 0.12338300049304962,
      "learning_rate": 0.0004246987951807229,
      "loss": 0.0451,
      "step": 2000
    },
    {
      "epoch": 3.765060240963855,
      "grad_norm": 0.08597979694604874,
      "learning_rate": 0.0004058734939759036,
      "loss": 0.0386,
      "step": 2500
    },
    {
      "epoch": 4.518072289156627,
      "grad_norm": 0.0988745242357254,
      "learning_rate": 0.00038704819277108433,
      "loss": 0.0352,
      "step": 3000
    },
    {
      "epoch": 5.271084337349397,
      "grad_norm": 0.11785969883203506,
      "learning_rate": 0.00036822289156626507,
      "loss": 0.0331,
      "step": 3500
    }
  ],
  "logging_steps": 500,
  "max_steps": 13280,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 2154066276777984.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}