Add eos_token_id to generation config

Currently, the model never stops generating because the eos_token_id is None inside HF generate function.
The eos_token_id is inferred from the generation config.
This token id can be passed in explicitely but its better to have it in the config.

generation_config.json

@@ -1,5 +1,6 @@
 {
   "_from_model_config": true,
   "transformers_version": "4.28.1",
+  "eos_token_id": 0,
   "use_cache": false
 }