Downloads · 30 days
16
13% of all-time downloads
AmanPriyanshu/gpt-oss-16.1b-specialized-law-pruned-moe-only-24-experts
gpt-oss-16.1b-specialized-law-pruned-moe-only-24-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
16
13% of all-time downloads
All-time downloads
123
Public
Parameters
16.1B
32.3 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors32.3 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 24 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 | ~16.1B (pruned from 21B) |
| Original Experts per Layer | 32 |
| Pruned Experts per Layer | 24 |
| 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 24 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-16.1b-specialized-law-pruned-moe-only-24-experts",
torch_dtype=torch.bfloat16,
device_map="cpu",
trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-16.1b-specialized-law-pruned-moe-only-24-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-16.1b-specialized-law-pruned-moe-only-24-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-16.1b-specialized-law-pruned-moe-only-24-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 the two main branches of the legal system, and they serve very different purposes. Below is a concise comparison that highlights the key distinctions:
| Feature | Civil Law | Criminal Law |
|--------|-----------|--------------|
| **Purpose** | To resolve disputes between private parties (individuals, businesses, organizations). | 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 / Remedy** | Usually **compensation** (damages) or **injunctive relief** (orders to do or stop doing something). | **Punishment** (e.g., fines, imprisonment, probation, community service) and sometimes restitution. |
| **Nature of Wrong** | Wrongful acts that cause loss or injury to another party (e.g., breach of contract, negligence, defamation). | Acts that violate criminal statutes (e.g., theft, assault, fraud, homicide). |
| **Procedure** | Often starts with a civil lawsuit filed in a civil court; parties may negotiate or go to trial. | Begins with an arrest or indictment; the case is prosecuted 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 prove innocence or rely on defenses. |
| **Appeals** | Appeals usually go to a higher civil court; the focus is on correcting legal errors. | Appeals can go to higher criminal courts; the focus is on correcting procedural or substantive errors. |
| **Examples** | Contract disputes, property disputes, personal injury claims, family law (divorce, custody). | Theft, assault, drug offenses, murder, tax evasion. |
### Key Take‑aways
1. **Who is suing?**
- *Civil*: A private party (individual or entity).
- *Criminal*: The state or government
Question: Explain the concept of due process in legal systems.
Response:
<|channel|>analysis<|message|>Need explanation.<|end|><|start|>assistant<|channel|>final<|message|>**Due Process – A Cornerstone of Fair Legal Systems**
---
### 1. What is Due Process?
*Due process* is a constitutional or statutory guarantee that the state will not deprive an individual of life, liberty, or property without a fair and orderly legal procedure. It is a safeguard against arbitrary or unjust government action. In the United States, it is enshrined in the Fifth and Fourteenth Amendments; in other jurisdictions it appears in constitutions, statutes, or common‑law principles.
---
### 2. Two Main Branches
| Branch | Key Elements | Typical Applications |
|--------|--------------|-----------------------|
| **Procedural Due Process** | • Notice of the action or charge<br>• Opportunity to be heard (e.g., trial, hearing)<br>• Decision by an impartial decision‑maker<br>• Reasoned decision (rationale) | Criminal prosecutions, civil lawsuits, administrative hearings, tax assessments, immigration removals |
| **Substantive Due Process** | • The law itself must be fair, reasonable, and not arbitrary<br>• Protects fundamental rights that are not explicitly listed in the constitution | Laws that affect fundamental rights (e.g., laws restricting free speech, laws that impose a heavy burden on a protected class) |
---
### 3. Core Components
| Component | Explanation | Example |
|-----------|-------------|---------|
| **Notice** | The person must be informed of the claim, charge, or action against them. | A defendant receives a summons and complaint. |
| **Opportunity to Respond** | The person must be able to present evidence, cross‑examine witnesses, and argue their case. | A civil plaintiff files a defense and presents evidence at trial. |
| **Impartial Decision‑Maker** | The judge, jury, or administrative body must be unbiased and free from conflicts of interest. | A judge recuses themselves if they have a personal stake. |
| **Reasoned Decision** | The decision must be based on evidence and legal reasoning, not arbitrary whims. | A court issues a written opinion explaining its findings. |
| **Right to Appeal** | The person can challenge a decision in a higher court. | A criminal defendant appeals a conviction to the appellate court. |
---
### 4. How Due Process Works in Practice
1. **Initiation**
- A government agency or private party initiates a legal action (e.g., filing a
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 freedoms that are considered fundamental to a democratic society. These rights are:
| **Right** | **Description** |
|-----------|-----------------|
| **Freedom of Religion** | Includes both the **free exercise** of religion (the right to practice any religion or none at all) and the **establishment clause** (prohibiting the government from establishing an official religion or favoring one religion over another). |
| **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 the right of the media to publish news, opinions, and investigative reporting without prior restraint by the government. |
| **Freedom of Assembly** | Allows people to gather peacefully for protests, demonstrations, or other collective activities. |
| **Freedom to Petition the Government** | Gives citizens the right to request changes in government policy or to seek redress for grievances. |
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}
}