:sparkles: initial commit

.github/workflows/push_to_hub.yml

@@ -0,0 +1,20 @@
+name: Sync to Hugging Face hub
+on:
+  push:
+    branches: [main]
+
+  # to run this workflow manually from the Actions tab
+  workflow_dispatch:
+
+jobs:
+  sync-to-hub:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
+          lfs: true
+      - name: Push to hub
+        env:
+          HF_TOKEN: ${{ secrets.HF_TOKEN }}
+        run: git push https://willsh1997:$HF_TOKEN@huggingface.co/spaces/willsh1997/knowledge-cutoff-gradio main

README.md

@@ -0,0 +1,14 @@
+---
+title: Knowledge Cutoff Gradio
+emoji: 🏆
+colorFrom: pink
+colorTo: yellow
+sdk: gradio
+sdk_version: 5.23.3
+app_file: knowledge_cutoff_demo.py
+pinned: false
+license: apache-2.0
+short_description: compare different llama versions for knowledge cutoff
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

knowledge_cutoff_demo.py

@@ -0,0 +1,156 @@
+import spaces
+from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
+import torch
+from transformers import pipeline
+import pandas as pd
+import gradio as gr
+
+#Llama 3.2 3b setup
+llama1_model_id = "huggyllama/llama-7b"
+llama1_pipe = pipeline(
+    "text-generation",
+    model=llama1_model_id,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
+#Llama 2 7b chat setup
+llama2_model_id = "meta-llama/Llama-2-7b-chat-hf"
+llama2_pipe = pipeline(
+    "text-generation",
+    model=llama2_model_id,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
+
+
+#Llama 3.2 3b setup
+llama3_model_id = "meta-llama/Llama-3.2-3B-Instruct"
+llama3_pipe = pipeline(
+    "text-generation",
+    model=llama3_model_id,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
+
+# #llama 4 setup
+# llama4_model_id = "meta-llama/Llama-4-Scout-17B-16E-Instruct"
+# llama4_pipe = pipeline(
+#     "text-generation",
+#     model=llama4_model_id,
+#     torch_dtype=torch.bfloat16,
+#     device_map="auto",
+# )
+
+########################
+# from transformers import AutoProcessor, Llama4ForConditionalGeneration
+# import torch
+
+# llama4_model_id = "meta-llama/Llama-4-Scout-17B-16E-Instruct"
+
+# llama4_processor = AutoProcessor.from_pretrained(llama4_model_id)
+# llama4_model = Llama4ForConditionalGeneration.from_pretrained(
+#     llama4_model_id,
+#     attn_implementation="flex_attention",
+#     device_map="auto",
+#     torch_dtype=torch.bfloat16,
+# )
+
+# def llama4_generate(input_question):
+#     messages = [
+#         {"role": "system", "content": [
+#             {"type": "text", "text": "You are a helpful chatbot assistant. Answer all questions in the language they are asked in."}
+#              ]
+#         },
+#         {"role": "user", "content": [
+#             {"type": "text", "text": input_question}
+#             ]
+#         },
+#         ]
+    
+#     inputs = llama4_processor.apply_chat_template(
+#         messages,
+#         add_generation_prompt=True,
+#         tokenize=True,
+#         return_dict=True,
+#         return_tensors="pt",
+#     ).to(model.device)
+    
+#     outputs = llama4_model.generate(
+#         **inputs,
+#         max_new_tokens=512,
+#     )
+    
+#     response = llama4_processor.batch_decode(outputs[:, inputs["input_ids"].shape[-1]:])[0]
+#     print(response)
+#     return response
+# print(outputs[0])
+
+#########################
+
+
+
+@spaces.GPU
+def llama_QA(input_question, pipe):
+    """
+    stupid func for asking llama a question and then getting an answer
+    inputs:
+    - input_question [str]: question for llama to answer
+    outputs:
+    - response [str]: llama's response
+    """
+    
+    messages = [
+    {"role": "system", "content": "You are a helpful chatbot assistant. Answer all questions in the language they are asked in."},
+    {"role": "user", "content": input_question},
+    ]
+    outputs = pipe(
+        messages,
+        max_new_tokens=512
+    )
+    response = outputs[0]["generated_text"][-1]['content']
+    return response
+
+
+@spaces.GPU
+def gradio_func(input_question, left_lang, right_lang):
+    """
+    silly wrapper function for gradio that turns all inputs into a single func. runs both the LHS and RHS of teh 'app' in order to let gradio work correctly.
+    """
+    output1 = llama_QA(input_question, llama1_pipe)
+    output2 = llama_QA(input_question, llama2_pipe)
+    output3 = llama_QA(input_question, llama3_pipe)
+    # output4 = llama4_generate(input_question)
+    return output1,output2,output3, #output4
+
+# Create the Gradio interface
+def create_interface():
+    
+    with gr.Blocks() as demo:
+        gr.Markdown("ask four different llama models the same question")
+        with gr.Row():
+            question_input = gr.Textbox(label="Enter your question", interactive=True)
+        with gr.Row():
+            submit_btn = gr.Button("Translate")
+        with gr.Row():
+            output1 = gr.Textbox(label="llama 1 output", interactive=False)
+            output2 = gr.Textbox(label="llama 2 output", interactive=False)
+            output3 = gr.Textbox(label="llama 3 output", interactive=False)
+            # output4 = gr.Textbox(label="llama 4 output", interactive=False)
+
+            
+        submit_btn.click(
+            fn=gradio_func,
+            inputs=[question_input],
+            outputs=[
+                    output1,
+                     output2,
+                     output3,
+                     # output4,
+                    ]
+        )
+    
+    return demo
+
+# Launch the app
+demo = create_interface()
+demo.launch()

requirements.txt

@@ -0,0 +1,87 @@
+accelerate==1.4.0
+aiofiles==23.2.1
+annotated-types==0.7.0
+anyio==4.8.0
+asttokens==3.0.0
+bitsandbytes==0.45.4
+certifi==2025.1.31
+charset-normalizer==3.4.1
+click==8.1.8
+comm==0.2.2
+debugpy==1.8.12
+decorator==5.1.1
+exceptiongroup==1.2.2
+executing==2.2.0
+fastapi==0.115.8
+ffmpy==0.5.0
+filelock==3.17.0
+fsspec==2025.2.0
+gradio==5.16.1
+gradio_client==1.7.0
+h11==0.14.0
+httpcore==1.0.7
+httpx==0.28.1
+huggingface-hub
+idna==3.10
+ipykernel==6.29.5
+ipython==8.32.0
+jedi==0.19.2
+Jinja2==3.1.5
+jupyter_client==8.6.3
+jupyter_core==5.7.2
+markdown-it-py==3.0.0
+MarkupSafe==2.1.5
+matplotlib-inline==0.1.7
+mdurl==0.1.2
+mpmath==1.3.0
+nest-asyncio==1.6.0
+networkx==3.4.2
+numpy==2.2.3
+orjson==3.10.15
+packaging==24.2
+pandas==2.2.3
+parso==0.8.4
+pexpect==4.9.0
+pillow==11.1.0
+platformdirs==4.3.6
+prompt_toolkit==3.0.50
+psutil==7.0.0
+ptyprocess==0.7.0
+pure_eval==0.2.3
+pydantic==2.10.6
+pydantic_core==2.27.2
+pydub==0.25.1
+Pygments==2.19.1
+python-dateutil==2.9.0.post0
+python-multipart==0.0.20
+pytz==2025.1
+PyYAML==6.0.2
+pyzmq==26.2.1
+regex==2024.11.6
+requests==2.32.3
+rich==13.9.4
+ruff==0.9.6
+safehttpx==0.1.6
+safetensors==0.5.2
+semantic-version==2.10.0
+shellingham==1.5.4
+six==1.17.0
+sniffio==1.3.1
+stack-data==0.6.3
+starlette==0.45.3
+sympy==1.13.1
+tokenizers==0.21.0
+tomlkit==0.13.2
+torch==2.4.0
+tornado==6.4.2
+tqdm==4.67.1
+traitlets==5.14.3
+transformers==4.51.0
+typer==0.15.1
+typing_extensions==4.12.2
+tzdata==2025.1
+urllib3==2.3.0
+uvicorn==0.34.0
+wcwidth==0.2.13
+websockets==14.2
+