Downloads · 30 days
17
13% of all-time downloads
AmanPriyanshu/gpt-oss-6.0b-specialized-law-pruned-moe-only-7-experts
gpt-oss-6.0b-specialized-law-pruned-moe-only-7-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
17
13% of all-time downloads
All-time downloads
136
Public
Parameters
6B
12 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors12 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 7 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 | ~6.0B (pruned from 21B) |
| Original Experts per Layer | 32 |
| Pruned Experts per Layer | 7 |
| 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 7 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-6.0b-specialized-law-pruned-moe-only-7-experts",
torch_dtype=torch.bfloat16,
device_map="cpu",
trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-6.0b-specialized-law-pruned-moe-only-7-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-6.0b-specialized-law-pruned-moe-only-7-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-6.0b-specialized-law-pruned-moe-only-7-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|>User asks: difference between civil and criminal law. Provide explanation.<|end|><|start|>assistant<|channel|>final<|message|>### Civil vs. Criminal Law
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Question: Explain the concept of due process in legal systems.
Response:
<|channel|>analysis<|message|>Need to explain due process concept.<|end|><|start|>assistant<|channel|>final<|message|>### What is “Due Process” (or “Due‑Process” in short)
In the legal system, **“due process”** is a principle that every person, whether a defendant, a victim, or a witness, is entitled to a fair, lawful, and just treatment under the law. It is not a single rule, but a set of norms that must be respected in every legal procedure. In short, it is the **“right to a fair hearing”**.
### Core Elements
1. **Notice of the Charge** – the person must be informed of the allegations or the charges they face.
2. **Right to Counsel** – the person has the right to hire an attorney or to have one appointed if they cannot afford one.
3. **Right to a Public Hearing** – the case must be heard in a public court or a hearing that is open to the public.
4. **Right to a Fair Jury** – the jury must be impartial, and the person must have a chance to challenge the evidence or the witnesses.
5. **Right to a Fair Sentive** – the person has the right to a fair sentencing, and the sentence must be proportionally appropriate.
6. **Right to a Fair Appeal** – the person has the right to appeal the decision or to have a review.
7. **Right to a Fair Release** – the person has the right to be released from custody or to be released from the prison.
8. **Right to a Fair Compensation** – the person has the right to compensation for any wrong or for any harm.
### Why It Matters
- **Fairness** – it ensures that the legal system is not arbitrary or that it is not biased.
- **Rule of Law** – it ensures that the law is applied consistently.
- **Human Rights** – it protects the human dignity and the human life.
### How It Is Implemented
- **Stat laws** – many statutes in the U.S. and other countries codize the right to due process.
- **Court rules** – judges and attorneys must follow the due process.
- **Judices** – the courts interpret the due process in a variety of cases.
### Examples
- **Mir‑Case** – the Supreme Court’s decision that a defendant has a right to a fair hearing.
- **Habe‑Case** – the case where
Question: What are the basic rights protected by the First Amendment?
Response:
<|channel|>analysis<|message|>User asks: "What are the basic rights protected by the First Amendment?" They want answer. Provide explanation.<|end|><|start|>assistant<|channel|>final<|message|>The First Amendment, as outlined in the U‑C‑Con (the U‑C‑Con is a constitution that is a constitution of the United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United United
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}
}