You can deploy this model and start earning money today!

AmanPriyanshu/
gpt-oss-13.7b-specialized-law-pruned-moe-only-20-experts

Name: gpt-oss-13.7b-specialized-law-pruned-moe-only-20-experts
Availability: InStock

Views

No views yet

Law GPT-OSS Model (20 Experts)

Project: https://amanpriyanshu.github.io/GPT-OSS-MoE-ExpertFingerprinting/

👥 Follow the Authors

Aman Priyanshu

Supriti Vijay

Introduction

This is a pruned variant of OpenAI's GPT-OSS-20B model, reduced to 20 experts per layer based on activation patterns from the AmanPriyanshu/GPT-OSS-20B MoE Expert Activations dataset. We analyzed router decisions across evaluation benchmarks to identify and retain experts most relevant for law tasks.

⚠️ Experimental Model: This is an experimental pruned model that may not work well - check the examples below to see if the outputs meet your needs before use.

This pruning approach reduces the model size while attempting to preserve performance on the target domain.

Model Architecture & Statistics

Metric	Value
Base Model	openai/gpt-oss-20b
Architecture	Mixture-of-Experts Transformer
Total Parameters	~13.7B (pruned from 21B)
Original Experts per Layer	32
Pruned Experts per Layer	20
Layers	24
Top-k Routing	4
Context Length	128K tokens
Attention Heads	64 (Query), 8 (Key-Value)
Residual Dimension	2880
Attention Pattern	Alternating dense & sliding window (128 tokens)
Positional Encoding	RoPE (Rotary Position Embedding)
Normalization	RMSNorm
Precision	BF16
License	Apache 2.0
Specialization	Law

Pruning Methodology

What is Expert Pruning?

Mixture-of-Experts models contain multiple specialized sub-networks (experts) per layer. During inference, only a subset of experts are activated for each token. Expert pruning involves:

Analyzing Usage Patterns: Tracking which experts activate most frequently for specific tasks
Removing Underutilized Experts: Discarding experts with low activation rates for the target domain
Preserving Router Functionality: Maintaining the routing mechanism with fewer available experts

Our Approach

Data-Driven Selection: Used activation patterns from law evaluation tasks
Systematic Reduction: Reduced from 32 to 20 experts per layer
No Retraining: Direct removal without additional training steps

Performance & Applications

Pruning Benefits

Smaller Memory Footprint: 62.5% of original expert parameters
Reduced Computational Load: Fewer routing decisions during inference
Focused Capabilities: Retains experts relevant to law tasks

Use Cases

Speculative Decoding: Draft model for full GPT-OSS-20B
Resource-Constrained Deployment: Edge devices, mobile applications
Research: Study expert specialization in MoE models
Fine-tuning: Smaller base model for domain adaptation

Note: Performance may vary depending on how well the pruned experts match your specific use case.

Motivation & Expert Selection

This legal domain model employs experts that demonstrated expertise during law-related tasks from MMLU legal subjects. These experts excel at legal reasoning, jurisprudence, and understanding of legal frameworks and procedures.

The expert selection process utilized our comprehensive analysis of router activation patterns across multiple evaluation benchmarks:

GPQA: Graduate-level questions in physics, chemistry, biology (Diamond & Expert subsets)
MMLU/MMLU-Pro: Comprehensive knowledge across 57+ subjects including science, medicine, law
SORRY-Bench: Safety evaluation across harmful content categories
Tulu3: Persona-driven instruction following with verifiable constraints
Polyglot-or-Not: Multilingual factual completion tasks

By identifying experts that consistently activated for law tasks, we created this specialized model that maintains domain expertise while significantly reducing computational requirements from 32 to 20 experts per layer.

Dataset & Analysis Foundation

This model is based on analysis from the GPT-OSS-20B MoE Expert Activations dataset available at: 🔗 https://huggingface.co/datasets/AmanPriyanshu/GPT-OSS-20B-MoE-expert-activations

The dataset contains router activation patterns from OpenAI's GPT-OSS-20B model across diverse evaluation benchmarks, enabling the creation of these domain-optimized models through systematic expert pruning.

Pruning Methodology

Our approach involves:

Activation Analysis: Comprehensive evaluation of expert usage patterns across domain-specific tasks
Expert Ranking: Identification of the most frequently activated experts for target domains
Systematic Pruning: Reduction from 32 to 20 experts while preserving router functionality
Quality Validation: Testing to ensure maintained performance on target tasks

This is a direct pruning approach - no additional training was performed. The model inherits all capabilities from the original GPT-OSS-20B with focused expert selection.

Usage

CPU Inference

python

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# Load the specialized model on CPU
model = AutoModelForCausalLM.from_pretrained(
    "AmanPriyanshu/gpt-oss-13.7b-specialized-law-pruned-moe-only-20-experts", 
    torch_dtype=torch.bfloat16, 
    device_map="cpu", 
    trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-13.7b-specialized-law-pruned-moe-only-20-experts")

# Generate with the model
messages = [
    {"role": "user", "content": "What is the difference between civil and criminal law?"}
]

inputs = tokenizer.apply_chat_template(
    messages, 
    add_generation_prompt=True, 
    return_tensors="pt", 
    return_dict=True,
    reasoning_effort="medium"
)

# Ensure inputs are on the same device as model
inputs = {k: v.to(model.device) for k, v in inputs.items()}

outputs = model.generate(
    **inputs, 
    max_new_tokens=512,
    do_sample=True,
    temperature=0.1,
    top_p=0.9,
    pad_token_id=tokenizer.eos_token_id,
    eos_token_id=tokenizer.eos_token_id
)

# Decode only the generated part
input_length = inputs['input_ids'].shape[1]
response_tokens = outputs[0][input_length:]
response = tokenizer.decode(response_tokens, skip_special_tokens=True)
print(response)

Apple Silicon (MPS) Inference

python

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# Check MPS availability and load model
device = "mps" if torch.backends.mps.is_available() else "cpu"

model = AutoModelForCausalLM.from_pretrained(
    "AmanPriyanshu/gpt-oss-13.7b-specialized-law-pruned-moe-only-20-experts", 
    torch_dtype=torch.float16,  # Better MPS compatibility
    device_map=device, 
    trust_remote_code=True,
    low_cpu_mem_usage=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-13.7b-specialized-law-pruned-moe-only-20-experts")

# Generate with the model
messages = [
    {"role": "user", "content": "What is the difference between civil and criminal law?"}
]

inputs = tokenizer.apply_chat_template(
    messages, 
    add_generation_prompt=True, 
    return_tensors="pt", 
    return_dict=True,
    reasoning_effort="medium"
)

# Move inputs to model device
inputs = {k: v.to(model.device) if isinstance(v, torch.Tensor) else v for k, v in inputs.items()}

# Use torch.no_grad for MPS stability
with torch.no_grad():
    outputs = model.generate(
        **inputs, 
        max_new_tokens=512,
        do_sample=True,
        temperature=0.1,
        top_p=0.9,
        pad_token_id=tokenizer.eos_token_id,
        eos_token_id=tokenizer.eos_token_id,
        use_cache=True
    )

# Decode only the generated part
input_length = inputs['input_ids'].shape[1]
response_tokens = outputs[0][input_length:]
response = tokenizer.decode(response_tokens, skip_special_tokens=True)
print(response)

GPU Inference

For CUDA GPU inference, the models work with minimal modifications from the CPU example above. Simply change:

python

device_map="auto"  # Will automatically use GPU if available
torch_dtype=torch.bfloat16  # or torch.float16

Model Examples

The following examples demonstrate the model's performance on law tasks:

Example 1

Question: What is the difference between civil and criminal law?

Response:

<|channel|>analysis<|message|>Need explain difference.<|end|><|start|>assistant<|channel|>final<|message|>**Civil law** and **criminal law** are the two main branches of the legal system that deal with very different kinds of disputes and wrongs.  While they share some procedural tools (e.g., courts, judges, evidence rules), their purposes, parties, burdens of proof, and consequences are distinct.

| Feature | Civil Law | Criminal Law |
|---|---|---|
| **Purpose** | Resolve disputes between private parties (individuals, businesses, or organizations). | Protect society and maintain public order by punishing conduct that is deemed harmful or dangerous. |
| **Parties** | Plaintiff (the person or entity that claims harm) vs. Defendant (the person or entity accused of wrongdoing). | State (or government) vs. Defendant (the person accused of a crime). |
| **Nature of Wrong** | Wrongful acts that cause injury, loss, or breach of contract. | Acts that violate statutes or criminal codes (e.g., theft, assault, murder). |
| **Burden of Proof** | **Pre‑ponderance of evidence** – “more likely than not.” | **Beyond a reasonable doubt** – the highest standard. |
| **Remedies** | Usually **compensation** (damages) or **specific performance** (e.g., injunction, contract enforcement). | **Punitive** (penalties such as fines, imprisonment, probation) and sometimes restitution. |
| **Procedures** | Often **civil litigation**: pleadings, discovery, trial, settlement. | **Criminal proceeding**: arraignment, plea, trial, sentencing. |
| **Rights of the Defendant** | Right to a fair trial, but no right to a jury in all jurisdictions (though many have jury trials). | Strong constitutional protections: right to counsel, right against self‑incrimination, right to a jury, right to a speedy trial. |
| **Outcome** | The court orders the defendant to pay damages, perform an act, or refrain from an act. | The court imposes a penalty (fine, imprisonment, community service, etc.). |
| **Appeals** | Usually to a higher civil court; the standard of review is often “deference to the trial court’s findings.” | Appeals are often to a higher criminal court; the appellate court reviews for legal errors, not factual findings. |

### Key Differences in Practice

1. **Who Initiates the Action?**  
   - *Civil

Example 2

Question: Explain the concept of due process in legal systems.

Response:

<|channel|>analysis<|message|>Need explain due process.<|end|><|start|>assistant<|channel|>final<|message|>**Due Process (or “Due Process”) – A Cornerstone of Fair Justice**

---

### 1. What is Due Process?

*Due process* is the set of legal safeguards that ensure a person’s rights are respected before the state takes any action that could affect that person’s liberty, property, or other legal interests. It is the procedural counterpart to substantive rights (e.g., the right to life, property, or freedom from discrimination). In short, *due process* guarantees that the government follows a fair, transparent, and predictable set of steps before making a decision that matters to you.

---

### 2. Core Elements

| Element | Meaning | Typical Example |
|---------|----------|----------------|
| **Notice** | The individual must be informed of the action, the allegations, and the consequences. | A summons to a criminal trial, a notice of eviction. |
| **Opportunity to Respond** | The person must be given a chance to present evidence, argue, and cross‑examine witnesses. | A hearing before a judge, a trial. |
| **Neutral Decision‑Maker** | The decision must be made by an impartial judge, jury, or administrative body. | A judge in a civil case, a jury in a criminal case. |
| **Right to Counsel** | The individual may have legal representation, especially in serious matters. | Defense counsel in criminal proceedings. |
| **Rule of Law** | The process must follow established statutes, regulations, and precedent. | Statutory criminal code, administrative regulations. |
| **Procedural Consistency** | Similar cases are treated similarly; the process is not arbitrary. | Equal treatment of all defendants in a criminal case. |
| **Judicial Review** | The decision can be challenged in a higher court. | Appeal of a conviction. |

---

### 3. Why It Matters

1. **Prevention of Arbitrary Power** – Without due process, governments could act capably and unpredictably, undermining liberty and property rights.  
2. **Transparency & Accountability** – The public can see how decisions are made, fostering trust in the legal system.  
3. **Error Correction** – Mistakes can be identified and rectified through appeals or reviews.  
4. **Fairness & Equality** – All individuals receive the same procedural protections, regardless of status.

---

### 4. How It Appears in Different Legal Systems

| System | Key Features of

Example 3

Question: What are the basic rights protected by the First Amendment?

Response:

<|channel|>analysis<|message|>Need to answer: freedom of religion, speech, press, assembly, petition.<|end|><|start|>assistant<|channel|>final<|message|>The First Amendment to the United States Constitution protects several core civil liberties that are considered essential to a free and democratic society. These rights are:

| **Right** | **Description** | **Key Points** |
|-----------|----------------|--------------|
| **Freedom of Religion** | Individuals may practice any religion, or none, without government interference. | Includes both the *establishment clause* (no state religion) and the *free‑exercise clause* (no undue restrictions on religious practice). |
| **Freedom of Speech** | Citizens can express ideas, opinions, and information without censorship. | Includes political speech, symbolic actions, and, in most cases, the right to criticize the government. |
| **Freedom of the Press** | Media outlets can publish news, opinions, and investigative reporting. | Protects the press from prior restraint and government control over content. |
| **Right to Assemble** | People may gather peacefully for protests, rallies, or other collective activities. | Includes the right to form and join associations, unions, or political parties. |
| **Right to Petition the Government** | Citizens can request changes to laws or policies and seek redress from the government. | Covers petitions, lobbying, and other forms of direct communication with officials. |

These five freedoms together form the foundation of the First Amendment’s protection of individual liberty and democratic participation.<|return|>

Citation

If you use this model in your research, please cite:

bibtex

@misc{priyanshu2025gptoss,
  title={{GPT-OSS MoE Expert Fingerprinting: Analyzing Expert Activation Patterns in Mixture of Experts Models}},
  author={Priyanshu, Aman and Vijay, Supriti},
  year={2025},
  howpublished={\url{https://amanpriyanshu.github.io/GPT-OSS-MoE-ExpertFingerprinting/}},
  note={Interactive analysis tool for expert activation patterns in MoE architectures}
}

References & Resources

Original Model: OpenAI GPT-OSS Model Card
Model Hub: GPT-OSS-20B on Hugging Face
Expert Analysis Dataset: GPT-OSS-20B MoE Expert Activations
Project Page: GPT-OSS MoE Expert Fingerprinting
GitHub Repository: OpenAI GPT-OSS