Upload folder using huggingface_hub

README.md

@@ -1,202 +1,95 @@
----
-base_model: unsloth/gemma-2-2b-bnb-4bit
-library_name: peft
----
-
-# Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
-
-
+# GPQA Generator: Fine-tuned Gemma 2B for Google-Proof Question Generation
 
 ## Model Details
 
-### Model Description
-
-<!-- Provide a longer summary of what this model is. -->
-
-
-
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-
-### Model Sources [optional]
-
-<!-- Provide the basic links for the model. -->
-
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-
-## Uses
-
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-
-### Direct Use
-
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-
-[More Information Needed]
-
-### Downstream Use [optional]
-
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-
-[More Information Needed]
-
-### Out-of-Scope Use
-
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-
-[More Information Needed]
-
-## Bias, Risks, and Limitations
-
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-
-[More Information Needed]
-
-### Recommendations
-
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-
-## How to Get Started with the Model
-
-Use the code below to get started with the model.
-
-[More Information Needed]
-
-## Training Details
-
-### Training Data
-
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-
-[More Information Needed]
-
-### Training Procedure
-
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-
-#### Preprocessing [optional]
-
-[More Information Needed]
-
-
-#### Training Hyperparameters
-
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-
-#### Speeds, Sizes, Times [optional]
-
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-
-[More Information Needed]
-
-## Evaluation
-
-<!-- This section describes the evaluation protocols and provides the results. -->
-
-### Testing Data, Factors & Metrics
-
-#### Testing Data
-
-<!-- This should link to a Dataset Card if possible. -->
-
-[More Information Needed]
-
-#### Factors
-
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-
-[More Information Needed]
-
-#### Metrics
-
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-
-[More Information Needed]
-
-### Results
-
-[More Information Needed]
-
-#### Summary
-
-
-
-## Model Examination [optional]
-
-<!-- Relevant interpretability work for the model goes here -->
-
-[More Information Needed]
-
-## Environmental Impact
-
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+- **Model Type:** Language Model
+- **Base Model:** unsloth/gemma-2-2b-bnb-4bit
+- **Fine-tuned by:** [Your Organization/Name]
+- **License:** [Specify the license]
 
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
+This model is a fine-tuned version of the Gemma 2B base model, specifically tailored to generate Google-Proof Question and Answer (GPQA) sets. It produces graduate-level, context-rich multiple-choice questions along with one correct answer, three incorrect answers, and an explanation.
 
-## Technical Specifications [optional]
+## Intended Use
 
-### Model Architecture and Objective
+This model is designed for educational content creators, assessment developers, and researchers who need to generate complex, Google-proof multiple-choice questions across various academic disciplines.
 
-[More Information Needed]
+### Primary Use Cases:
 
-### Compute Infrastructure
+- Generating challenging assessment questions for advanced students
+- Creating content for educational platforms and applications
+- Assisting in the development of standardized tests
+- Supporting research in question generation and educational assessment
 
-[More Information Needed]
+## How to Use
 
-#### Hardware
+### Setting Up
 
-[More Information Needed]
+1. Clone the repository containing the model and scripts.
+2. Ensure you have the required dependencies installed (httpx, transformers, etc.).
 
-#### Software
+### Running the Model
 
-[More Information Needed]
+1. Start the vLLM server:
+   ```
+   ./run_vllm_2b.sh
+   ```
 
-## Citation [optional]
+2. Generate questions using the `generate.py` script:
+   
+   For a single category:
+   ```
+   python generate.py --category "Your Category" --depth 4
+   ```
 
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+   To use predefined categories:
+   ```
+   python generate.py --use-array --depth 4
+   ```
 
-**BibTeX:**
+### Configuration
 
-[More Information Needed]
+- Modify the `CATEGORIES_TO_PROCESS` list in the script to add or change predefined categories.
+- Adjust the `max_depth` parameter to control the depth of subcategory exploration.
+- The script uses multi-threading for efficient processing. Adjust `num_threads` in `process_categories()` if needed.
 
-**APA:**
+## Sample Output
 
-[More Information Needed]
+Here's an example of the generated output:
 
-## Glossary [optional]
+```json
+{
+  "question": "A developer is working on a large project that uses Mercurial version control. They need to merge a branch containing bug fixes from another team. What is the recommended approach to avoid merging conflicts?",
+  "answer": "The developer should: \n1. Create a new branch from the source directory. \n2. Compare the history of the branches. \n3. Resolve conflicts manually. \n4. Commit the changes. ",
+  "incorrect_answer_1": "Merge the branches directly.",
+  "incorrect_answer_2": "Skip the merge process altogether.",
+  "incorrect_answer_3": "Use a third-party tool like Git.",
+  "explanation": "Merging branches in Mercurial requires careful consideration to avoid conflicts. Here's a breakdown of the reasoning: \n1. Branch Creation: Creating a new branch allows the developer to isolate the changes from the other team without affecting the base branch.\n2. Conflict Detection: Comparing the histories of the branches helps identify potential conflicts that may arise during the merge.\n3. Conflict Resolution: Manual conflict resolution is essential to ensure the merge is successful. Mercurial provides tools like \"diff\" and \"merge\" commands for this purpose.\n4. Committing Changes: Once the merge is complete, the developer should commit the changes to their new branch.",
+  "subcategories": ["Version Control", "Mercurial", "Merge Conflicts"],
+  "category": "Mercurial",
+  "depth": 0
+}
+```
 
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+## Limitations
 
-[More Information Needed]
+- The model generates questions based on its training data, which may not always reflect the most current information in rapidly evolving fields.
+- While designed to be "Google-proof," the effectiveness may vary depending on the specific topic and how information is presented online.
+- The quality and accuracy of generated questions should be reviewed by subject matter experts before use in formal assessments.
 
-## More Information [optional]
+## Ethical Considerations
 
-[More Information Needed]
+- Users should be aware of potential biases in the generated content and review questions for fairness and inclusivity.
+- The model should not be used to generate misleading or factually incorrect information.
+- Respect copyright and intellectual property rights when using generated content.
 
-## Model Card Authors [optional]
+## Citation
 
-[More Information Needed]
+If you use this model in your research or applications, please cite it as follows:
 
-## Model Card Contact
+```
+[Citation information to be added]
+```
 
-[More Information Needed]
-### Framework versions
+## Contact
 
-- PEFT 0.12.0
+For questions, feedback, or support, please contact [Your Contact Information].

run_vllm_2b.sh

@@ -5,6 +5,7 @@ VLLM_ALLOW_RUNTIME_LORA_UPDATING=1 VLLM_ALLOW_LONG_MAX_MODEL_LEN=1 vllm serve ~/
   --trust-remote-code \
   --max-model-len 8192 \
   --disable-log-requests \
-  --guided-decoding-backend lm-format-enforcer \
   --enable-lora \
   --lora-modules gpqa=./gpqa \
+
+  # --guided-decoding-backend lm-format-enforcer \