Downloads · 30 days
15
11% of all-time downloads
AmanPriyanshu/gpt-oss-18.5b-specialized-law-pruned-moe-only-28-experts
gpt-oss-18.5b-specialized-law-pruned-moe-only-28-experts is a text generation model from AmanPriyanshu. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
Project: https://amanpriyanshu.github.io/GPT-OSS-MoE-ExpertFingerprinting/
Downloads · 30 days
15
11% of all-time downloads
All-time downloads
138
Public
Parameters
18.5B
37.1 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors37 GB · 100%
From the Hugging Face model README
Project: https://amanpriyanshu.github.io/GPT-OSS-MoE-ExpertFingerprinting/
<div align="center">This is a pruned variant of OpenAI's GPT-OSS-20B model, reduced to 28 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.
| Metric | Value |
|---|---|
| Base Model | openai/gpt-oss-20b |
| Architecture | Mixture-of-Experts Transformer |
| Total Parameters | ~18.5B (pruned from 21B) |
| Original Experts per Layer | 32 |
| Pruned Experts per Layer | 28 |
| 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 |
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:
Note: Performance may vary depending on how well the pruned experts match your specific use case.
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:
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 28 experts per layer.
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.
Our approach involves:
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.
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
# Load the specialized model on CPU
model = AutoModelForCausalLM.from_pretrained(
"AmanPriyanshu/gpt-oss-18.5b-specialized-law-pruned-moe-only-28-experts",
torch_dtype=torch.bfloat16,
device_map="cpu",
trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-18.5b-specialized-law-pruned-moe-only-28-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)
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-18.5b-specialized-law-pruned-moe-only-28-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-18.5b-specialized-law-pruned-moe-only-28-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)
For CUDA GPU inference, the models work with minimal modifications from the CPU example above. Simply change:
device_map="auto" # Will automatically use GPU if available
torch_dtype=torch.bfloat16 # or torch.float16
The following examples demonstrate the model's performance on law tasks:
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 two distinct branches of the legal system that serve different purposes, involve different parties, and have different procedures and outcomes. Below is a concise comparison of the key differences:
| Feature | Civil Law | Criminal Law |
|---------|-----------|--------------|
| **Purpose** | To resolve disputes between private parties (individuals, businesses, organizations) and to compensate the injured party. | To protect society by punishing conduct that is deemed harmful or dangerous to the public. |
| **Parties** | Plaintiff (the party bringing the claim) vs. Defendant (the party being sued). | State or government (prosecution) vs. Defendant (the accused). |
| **Burden of Proof** | **Preponderance of the evidence** – more likely than not (≈51%+). | **Beyond a reasonable doubt** – a much higher standard. |
| **Outcome** | Usually monetary damages, injunctions, or specific performance (ordering a party to do or not do something). | Penalties such as fines, probation, community service, imprisonment, or even death (in jurisdictions that allow it). |
| **Nature of Wrong** | Wrongful acts that cause injury or loss (e.g., breach of contract, negligence, defamation). | Acts that violate criminal statutes (e.g., theft, assault, murder). |
| **Procedure** | Generally starts with a civil complaint filed in a civil court; parties present evidence and arguments. | Begins with an arrest or formal charge; the prosecution presents evidence in a criminal court. |
| **Rights of the Defendant** | Right to a fair trial, but the defendant is not presumed guilty. | Presumption of innocence; the defendant must be proven guilty. |
| **Appeals** | Both parties can appeal, but the focus is on correcting legal errors that affected the outcome. | Appeals focus on correcting legal errors, procedural mistakes, or issues of sentencing. |
| **Evidence Rules** | More flexible; hearsay may be admissible if it is relevant. | Strict rules; hearsay is generally inadmissible unless it falls under an exception. |
| **Time Limits (Statutes of Limitations)** | Varies by claim type; often 1–10 years. | Generally no statute of limitations for serious crimes (e.g., murder), but some crimes have limits. |
| **Examples** | Contract disputes, property disputes, family law (
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** is a foundational principle that guarantees fair treatment through the normal judicial system, especially as a citizen’s rights are protected. It is a constitutional safeguard found in many democratic legal systems, most notably in the United States under the Fifth and Fourteenth Amendments, and in other countries through their own constitutions or human‑rights instruments.
---
## 1. What Does “Due Process” Mean?
At its core, due process requires that the government follow **procedural rules** before depriving a person of life, liberty, or property. It is not a single rule but a set of procedural guarantees that ensure:
1. **Notice** – The individual must be informed of the claims, charges, or actions against them.
2. **Opportunity to be Heard** – The individual must have a chance to present their side, challenge evidence, and cross‑examine witnesses.
3. **Impartial Decision‑Maker** – The decision must be made by an unbiased judge or tribunal.
4. **Reasoned Decision** – The outcome must be based on evidence and legal reasoning, not arbitrary discretion.
In many jurisdictions, due process is divided into two complementary strands:
| **Strand** | **Focus** | **Typical Requirements** |
|------------|-----------|--------------------------|
| **Procedural Due Process** | Fairness of the process | Notice, hearing, evidence, appeal |
| **Substantive Due Process** | Fairness of the law itself | Laws must not be arbitrary, discriminatory, or violate fundamental rights |
---
## 2. Historical Roots
- **English Common Law**: The Magna Carta (1215) and the Bill of Rights (1689) established early notions of fair trial and protection against arbitrary arrest.
- **American Constitution**: The Fifth Amendment (1791) and the Fourteenth Amendment (1868) incorporated due process into federal and state law.
- **International Law**: The Universal Declaration of Human Rights (1945) and the International Covenant on Civil and Political Rights (1960) embed due process in global norms.
---
## 3. Key Elements in Practice
| Element | Practical Example |
|---------|-------------------|
| **Notice** | A summons to a civil lawsuit, a notice of criminal charges, or a notice of a regulatory penalty. |
| **Hearing** | A trial before a judge or jury, a hearing before an administrative agency, or a mediation session. |
| **
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 fundamental to a free and democratic society. These rights are:
| Right | Key Elements |
|-------|--------------|
| **Freedom of Religion** | - **Establishment Clause**: The government may not establish an official religion or favor one religion over another. <br> - **Free‑Exercise Clause**: Individuals are free to practice any religion (or none) without government interference. |
| **Freedom of Speech** | - Individuals may express opinions, ideas, and information without prior government censorship. <br> - Includes political speech, artistic expression, and even offensive or unpopular viewpoints, subject to limited restrictions (e.g., incitement, defamation, obscenity). |
| **Freedom of the Press** | - The press (newspapers, books, radio, television, online media) can publish news and opinions without government control or prior restraint. |
| **Freedom of Assembly** | - People may gather peacefully for protests, demonstrations, or other collective activities. <br> - Requires that the government cannot prohibit peaceful assembly without a compelling reason. |
| **Right to Petition the Government** | - Citizens can formally request changes, express grievances, or seek redress from the government. <br> - Includes the right to file petitions, complaints, or appeals. |
These protections are not absolute; the Supreme Court has carved out narrow exceptions (e.g., for national security, public safety, or to prevent direct harm). Nonetheless, the First Amendment remains the cornerstone of American civil liberties, ensuring that individuals can practice their faith, speak freely, access information, gather peacefully, and seek redress from the state.<|return|>
If you use this model in your research, please cite:
@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}
}