Update app.py

app.py

@@ -70,7 +70,7 @@ def evaluate(
     for i in range(int(token_count)):
 
         input_ids = pipeline_v6.encode(ctx)[-ctx_limit:] if i == 0 else [token]
-        out, state = model_v6.forward(tokens=input_ids, state=state)
+        out, state = model_v6.forward(input_ids, state)
         for n in occurrence:
             out[n] -= (args.alpha_presence + occurrence[n] * args.alpha_frequency)
 
@@ -124,7 +124,7 @@ examples = [
 
 ##################################################################################################################
 with gr.Blocks(title=title_v6) as demo:
-    gr.HTML(f"<div style=\"text-align: center;\">\n<h1>{title_v6} (only 0.1B !!!)</h1>\n</div>")
+    gr.HTML(f"<div style=\"text-align: center;\">\n<h1>{title_v6} (!!! only 0.1B !!!)</h1>\n</div>")
 
     with gr.Tab("=== Base Model (Raw Generation) ==="):
         gr.Markdown(f"This is [RWKV-7 World v2.8](https://huggingface.co/BlinkDL/rwkv-7-world) 0.1B (L12-D768) - a 100% attention-free RNN [RWKV-LM](https://github.com/BlinkDL/RWKV-LM). Supports 100+ world languages and code. Check [400+ Github RWKV projects](https://github.com/search?o=desc&p=1&q=rwkv&s=updated&type=Repositories). *** Can try examples (bottom of page) *** (can edit them). Demo limited to ctxlen {ctx_limit}.")