checkpoint-5576/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 1000,
  "global_step": 5576,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003586800573888092,
      "grad_norm": 6.962954521179199,
      "learning_rate": 0.0,
      "loss": 4.6558,
      "step": 1
    },
    {
      "epoch": 0.3586800573888092,
      "grad_norm": 4.403517723083496,
      "learning_rate": 1.4332855093256816e-05,
      "loss": 3.7649,
      "step": 1000
    },
    {
      "epoch": 0.3586800573888092,
      "eval_cosine_accuracy": 0.9522557854652405,
      "eval_loss": 1.848149299621582,
      "eval_runtime": 21.5681,
      "eval_samples_per_second": 440.882,
      "eval_steps_per_second": 1.762,
      "step": 1000
    },
    {
      "epoch": 0.7173601147776184,
      "grad_norm": 4.781370639801025,
      "learning_rate": 1.9035549179021205e-05,
      "loss": 2.929,
      "step": 2000
    },
    {
      "epoch": 0.7173601147776184,
      "eval_cosine_accuracy": 0.9615101218223572,
      "eval_loss": 1.7497695684432983,
      "eval_runtime": 21.6499,
      "eval_samples_per_second": 439.218,
      "eval_steps_per_second": 1.755,
      "step": 2000
    },
    {
      "epoch": 1.0760401721664274,
      "grad_norm": 4.634256839752197,
      "learning_rate": 1.7441415590626496e-05,
      "loss": 2.5761,
      "step": 3000
    },
    {
      "epoch": 1.0760401721664274,
      "eval_cosine_accuracy": 0.963718593120575,
      "eval_loss": 1.6966071128845215,
      "eval_runtime": 21.8321,
      "eval_samples_per_second": 435.551,
      "eval_steps_per_second": 1.741,
      "step": 3000
    },
    {
      "epoch": 1.4347202295552366,
      "grad_norm": 4.340166091918945,
      "learning_rate": 1.5847282002231787e-05,
      "loss": 2.388,
      "step": 4000
    },
    {
      "epoch": 1.4347202295552366,
      "eval_cosine_accuracy": 0.9657166600227356,
      "eval_loss": 1.6643705368041992,
      "eval_runtime": 22.0348,
      "eval_samples_per_second": 431.545,
      "eval_steps_per_second": 1.725,
      "step": 4000
    },
    {
      "epoch": 1.793400286944046,
      "grad_norm": 4.487511157989502,
      "learning_rate": 1.4254742547425475e-05,
      "loss": 2.2845,
      "step": 5000
    },
    {
      "epoch": 1.793400286944046,
      "eval_cosine_accuracy": 0.9667683243751526,
      "eval_loss": 1.647828459739685,
      "eval_runtime": 21.6523,
      "eval_samples_per_second": 439.168,
      "eval_steps_per_second": 1.755,
      "step": 5000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 13940,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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
}