last touches

generate.py

@@ -22,8 +22,8 @@ logger = logging.getLogger(__name__)
 
 
 logger.warning("Loading model...")
-# model_id = "google/gemma-2b-it"
-model_id = "Qwen/Qwen1.5-0.5B-Chat"
+model_id = "google/gemma-2b-it"
+# model_id = "Qwen/Qwen1.5-0.5B-Chat"
 if torch.backends.mps.is_available():
     device = "mps"
     model = models.transformers(model_id, device=device)

gradio_app.py

@@ -6,7 +6,7 @@ import io
 import pandas as pd
 import spaces
 
-from generate import stream_jsonl_file
+from generate import model_id, stream_jsonl_file
 
 MAX_SIZE = 20
 DEFAULT_SEED = 42
@@ -58,7 +58,7 @@ def stream_more_output(query: str):
 
 title = "LLM DataGen"
 description = (
-    "Generate and stream synthetic dataset files in `{JSON Lines}` format\n\n"
+    f"Generate and stream synthetic dataset files in `{{JSON Lines}}` format (currently using [{model_id}](https://huggingface.co/{model_id}))\n\n"
     "Disclaimer: LLM data generation is an area of active research with known problems such as biased generation and incorrect information."
 )
 examples = [