Update generation_config.json

Pull in upstream second stop token.

Fixes issue where inference does not stop.

See upstream: https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct/blob/main/generation_config.json

generation_config.json

@@ -2,6 +2,6 @@
   "_from_model_config": true,
   "bos_token_id": 128000,
   "do_sample": true,
-  "eos_token_id": 128001,
+  "eos_token_id": [128001, 128009],
   "transformers_version": "4.40.0"
 }