Downloads · 30 days
11
9% of all-time downloads
AmanPriyanshu/gpt-oss-16.7b-specialized-law-pruned-moe-only-25-experts
gpt-oss-16.7b-specialized-law-pruned-moe-only-25-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
11
9% of all-time downloads
All-time downloads
124
Public
Parameters
16.7B
33.5 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors33.5 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 25 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 | ~16.7B (pruned from 21B) |
| Original Experts per Layer | 32 |
| Pruned Experts per Layer | 25 |
| 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 25 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-16.7b-specialized-law-pruned-moe-only-25-experts",
torch_dtype=torch.bfloat16,
device_map="cpu",
trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-16.7b-specialized-law-pruned-moe-only-25-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-16.7b-specialized-law-pruned-moe-only-25-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-16.7b-specialized-law-pruned-moe-only-25-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 to 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 that highlights 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** | **Pre‑ponderance of evidence** – the plaintiff must show that it is more likely than not that the defendant is liable. | **Beyond a reasonable doubt** – the prosecution must prove the defendant’s guilt to a very high standard. |
| **Outcome** | Usually monetary damages, injunctions, or specific performance (e.g., a contract must be fulfilled). | Penalties such as fines, probation, community service, or imprisonment. |
| **Procedure** | Civil courts (e.g., state or federal civil courts) handle claims; parties often represent themselves or hire attorneys. | Criminal courts (e.g., district or federal criminal courts) handle prosecutions; the state’s attorney represents the government. |
| **Rights of the Defendant** | Right to a fair trial, but the defendant is not presumed guilty. | Presumption of innocence; the defendant has the right to a jury, counsel, and to confront witnesses. |
| **Evidence Rules** | Generally more flexible; hearsay may be admissible if it is relevant. | Strict rules; hearsay is usually inadmissible unless an exception applies. |
| **Appeals** | Parties can appeal a civil judgment to a higher court. | The state can appeal a criminal conviction or sentence; the defendant can appeal a conviction or sentence. |
| **Typical Cases** | Breach of contract, property disputes, torts (e.g., negligence), family law (divorce, child custody). | Theft, assault, murder, drug offenses, fraud, white‑collar crimes. |
| **Statutory Basis** | Often governed by statutes, case law, and civil codes. | Governed by criminal statutes, the U.S.
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 under the law. It is a constitutional safeguard found in many legal systems—most notably in the U.S. Constitution (5th and 14th Amendments) and in the European Convention on Human Rights (Article 6). While the exact wording and scope can vary from country to country, the core idea is the same: the state must follow established, fair procedures before depriving a person of life, liberty, property, or other fundamental rights.
---
## 1. Two Faces of Due Process
| **Substantive Due Process** | **Procedural Due Process** |
|------------------------------|----------------------------|
| Protects the *content* of laws. | Protects the *process* by which laws are applied. |
| Requires that laws themselves be fair, reasonable, and not arbitrary. | Requires that the government follow fair procedures before taking action. |
| Example: A law that arbitrarily discriminates against a protected class violates substantive due process. | Example: A person cannot be imprisoned without a fair trial. |
In practice, the two are intertwined: a fair process is needed to enforce a fair law, and a fair law is needed to justify a fair process.
---
## 2. Core Elements of Procedural Due Process
| Element | What it Means | Typical Requirements |
|---------|---------------|-----------------------|
| **Notice** | The person must be informed of the action and the reasons behind it. | Written notice, clear description of charges, time to respond. |
| **Opportunity to be Heard** | The person must have a chance to present their side. | Hearing before a neutral decision‑maker, ability to present evidence, cross‑examine witnesses. |
| **Neutral Decision‑Maker** | The decision must be made by an impartial authority. | Judges, administrative officials, or other bodies with no bias. |
| **Reasoned Decision** | The decision must be based on evidence and legal reasoning. | Written opinions, record of evidence considered. |
| **Right to Appeal** | The person can challenge the decision in a higher forum. | Appeals process, appellate courts, administrative review. |
These elements are not rigid; courts often weigh them against the seriousness of the deprivation and the context. For example, a brief administrative hearing may be sufficient for a minor penalty, while a full trial is required for imprisonment.
---
## 3. How Due Process Works
Question: What are the basic rights protected by the First Amendment?
Response:
<|channel|>analysis<|message|>Need to answer: basic rights: 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** | **Description** |
|-----------|-----------------|
| **Freedom of Religion** | Includes both the *free exercise* of religion (the right to practice any faith or none at all) and the *establishment clause* (prohibits the government from endorsing or favoring a particular religion). |
| **Freedom of Speech** | Protects the right to express ideas, opinions, and information without government censorship or punishment. This includes spoken, written, and symbolic expression. |
| **Freedom of the Press** | Guarantees that newspapers, books, radio, television, and other media can publish news and opinions without prior restraint by the government. |
| **Freedom of Assembly** | Allows people to gather peacefully for protests, demonstrations, or other collective activities. |
| **Right to Petition the Government** | Gives citizens the right to appeal to the government for redress of grievances, including the right to file petitions, complaints, or appeals. |
These five freedoms together form the core of the First Amendment’s protection of individual liberty and democratic participation.<|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}
}