ucalyptus
/

prem-1B-grpo

@@ -1,199 +1,103 @@
 ---
 library_name: transformers
-tags: []
 ---
-# Model Card for Model ID
-<!-- Provide a quick summary of what the model is/does. -->
-## Model Details
-### Model Description
-<!-- Provide a longer summary of what this model is. -->
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
-- **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).
-- **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]
-## Technical Specifications [optional]
-### Model Architecture and Objective
-[More Information Needed]
-### Compute Infrastructure
-[More Information Needed]
-#### Hardware
-[More Information Needed]
-#### Software
-[More Information Needed]
-## Citation [optional]
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-**BibTeX:**
-[More Information Needed]
-**APA:**
-[More Information Needed]
-## Glossary [optional]
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-[More Information Needed]
-## More Information [optional]
-[More Information Needed]
-## Model Card Authors [optional]
-[More Information Needed]
-## Model Card Contact
-[More Information Needed]

 ---
+language:
+  - en
+tags:
+  - math
+  - reasoning
+  - grpo
+  - gsm8k
+  - reinforcement-learning
+license: apache-2.0
+datasets:
+  - openai/gsm8k
+metrics:
+  - accuracy
+  - format_adherence
 library_name: transformers
+pipeline_tag: text-generation
+base_model: premai-io/prem-1B-chat
 ---
+# GRPO-Trained Math Reasoning Model
+## Model Description
+This model was fine-tuned using GRPO (Generative Reward-Powered Optimization), a reinforcement learning technique that optimizes language models using multiple reward functions to improve both output format consistency and mathematical reasoning abilities.
+### Base Model
+- Started with the `premai-io/prem-1B-chat` base model
+- Uses Flash Attention 2 for efficient training
+- Model architecture: Causal Language Model (CLM)
+### GRPO Training Details
+GRPO training involves optimizing the model using two specific reward functions:
+1. **Format Reward Function**
+   - Ensures responses follow a strict XML-style format:
+   ```xml
+   <reasoning>
+   [step-by-step solution]
+   </reasoning>
+   <answer>
+   [numerical answer]
+   </answer>
+   ```
+   - Rewards are given for:
+     - Strict format adherence (0.5 points)
+     - Soft format matching (0.3 points)
+     - Integer answer format (0.5 points)
+     - Proper XML structure (up to 0.5 points)
+2. **Correctness Reward Function**
+   - Evaluates mathematical accuracy
+   - Awards 2.0 points for correct numerical answers
+   - Awards 0.0 points for incorrect answers
+### Training Configuration
+- Learning rate: 5e-6
+- Batch size: 2 per device
+- Gradient accumulation steps: 2
+- Training epochs: 1
+- Uses cosine learning rate scheduler
+- Warmup ratio: 0.1
+- Uses bfloat16 precision
+- Maximum prompt length: 256 tokens
+- Maximum completion length: 200 tokens
+- Number of generations per prompt: 16
+### Dataset
+- Trained on the GSM8K (Grade School Math 8K) dataset
+- Dataset contains grade-school level math word problems
+- Each problem includes a question and step-by-step solution
+## Intended Use
+- Solving mathematical word problems
+- Providing step-by-step reasoning for solutions
+- Educational assistance and math tutoring
+## Limitations
+- Limited to the complexity level of GSM8K problems
+- May struggle with problems requiring knowledge beyond the training data
+- Performance depends on proper formatting of input queries
+## Training Infrastructure
+- Supports distributed training across multiple GPUs
+- Uses NCCL backend for distributed processing
+- Implements gradient clipping (max norm: 0.1)
+## Evaluation
+The model's performance is continuously evaluated during training based on:
+1. Format adherence to the XML structure
+2. Mathematical accuracy of answers
+3. Quality of step-by-step reasoning
+## Citation
+If you use this model, please cite both the original GRPO paper and the GSM8K dataset.
 ## Training Details
 ### Training Data
+The model was trained on the [GSM8K dataset](https://huggingface.co/datasets/openai/gsm8k), which contains 8.5K grade school math word problems.
 ### Training Procedure
+- **Hardware:** Multi-GPU setup with NVIDIA GPUs
+- **Framework:** Hugging Face Transformers, TRL (Transformer Reinforcement Learning)
+- **Optimization:** GRPO with dual reward functions