Downloads · 30 days
11
9% of all-time downloads
AmanPriyanshu/gpt-oss-7.2b-specialized-instruction_following-pruned-moe-only-9-experts
gpt-oss-7.2b-specialized-instruction_following-pruned-moe-only-9-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
118
Public
Parameters
7.2B
14.4 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors14.3 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 9 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 instruction following 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 | ~7.2B (pruned from 21B) |
| Original Experts per Layer | 32 |
| Pruned Experts per Layer | 9 |
| 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 | Instruction Following |
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 instruction-following model leverages experts that excelled at constraint satisfaction tasks from Tulu3 Persona Instruction Following dataset. These experts specialize in precise adherence to user specifications and formatting requirements.
The expert selection process utilized our comprehensive analysis of router activation patterns across multiple evaluation benchmarks:
By identifying experts that consistently activated for instruction following tasks, we created this specialized model that maintains domain expertise while significantly reducing computational requirements from 32 to 9 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-7.2b-specialized-instruction_following-pruned-moe-only-9-experts",
torch_dtype=torch.bfloat16,
device_map="cpu",
trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-7.2b-specialized-instruction_following-pruned-moe-only-9-experts")
# Generate with the model
messages = [
{"role": "user", "content": "Write a formal email to a professor requesting a meeting, including: subject line, greeting, purpose, proposed times, and professional closing."}
]
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-7.2b-specialized-instruction_following-pruned-moe-only-9-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-7.2b-specialized-instruction_following-pruned-moe-only-9-experts")
# Generate with the model
messages = [
{"role": "user", "content": "Write a formal email to a professor requesting a meeting, including: subject line, greeting, purpose, proposed times, and professional closing."}
]
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 instruction following tasks:
Question: Write a formal email to a professor requesting a meeting, including: subject line, greeting, purpose, proposed times, and professional closing.
Response:
<|channel|>analysis<|message|>Need to produce formal email.<|end|><|start|>assistant<|channel|>final<|message|>**Subject:** Request for Meeting – Availability Inquiry
Dear Professor [Last Name],
I hope you are doing well. I am writing to request a brief meeting to discuss the progress of my research project and to seek your guidance on upcoming milestones.
Could you please let me know a convenient time for the meeting? I would be available on the following days:
- **Monday, 2024‑08‑13 – 30 – 30**
- **Tuesday, 2024‑08‑14 – 30 – 30**
- **Wednesday, 2024‑08‑15 – 30 – 30**
If any other times are more suitable, kindly let me know, and I will adjust accordingly.
Thank you for your time and consideration. I look forward to speaking with you.
Sincerely,
[Your Full Name]
[Your Program/Department]
[University]
[Phone] | [Email]
---
Feel free to adjust the times and contact details to fit your schedule.<|return|>
Question: Create a numbered list of steps for making scrambled eggs, using exactly 6 steps.
Response:
<|channel|>analysis<|message|>We need to give a numbered list of steps for making scrambled eggs, using exactly 6 steps. So we need 6 steps, numbered. Probably 1- 2- etc. Ensure exactly 6 steps. Let's produce 1- 2- 3- 4- 5- 6.<|end|><|start|>assistant<|channel|>final<|message|>**Steps to Make Scrambled Eggs (6 Steps)**
1. **Gather the ingredients** – Grab a few eggs (e.g., 3‑or‑4 eggs) and a small amount of butter or a light‑weight sauce.
2. **Open the eggs** – Crack each egg and let the yolc‑and‑white mix together.
3. **Heat the pan** – Turn the stove or a small‑sheat‑to‑heat the pan to a moderate temperature.
4. **Add the eggs to the pan** – Place the eggs in the pan and let them sit for a short amount of time.
5. **Let the eggs mix** – Allow the eggs to mix together, forming a smooth, creamy texture.
6. **Serve** – Take the eggs out of the pan, put them on a plate, and enjoy the finished scrambled eggs.<|return|>
Question: Summarize the water cycle in exactly 50 words, no more, no less.
Response:
<|channel|>analysis<|message|>Need 50 words exactly. Let's craft: "Water enters the atmosphere as vapor, condens into droplets, flows through rivers, oceans, and clouds, evaporates into air, returns to atmosphere, recondens, and repeats. The cycle of birth, flow, and death sustizes life, shaping ecosystems, nourishing bodies, and sustaining the planet." Count words: 1) Water(1) enters(2) the(3) atmosphere(4) as(5) vapor,(6) condens(7) into(8) droplets,(9) flows(10) through(11) rivers,(12) oceans,(13) and(14) clouds,(15) evaporates(16) into(20) air,(21) returns(22) to(23) atmosphere,(24) recondens,(25) and(26) repeats.(27) The(28) cycle(29) of(30) birth,(31) flow,(30) and(31) death (30) sustizes(30) life,(30) shaping(30) ecosystems,(30) nourishing(30) bodies,(30) and(30) sustaining(30) the(30) planet.(30) Need 50 words. Count: 1-1? Let's count properly: Water(1) enters(2) the(3) atmosphere(4) as(5) vapor,(6) condens(7) into(8) droplets,(9) flows(10) through(11) rivers,(12) oceans,(13) and(14) clouds,(15) evaporates(16) into(20) air,(21) returns(22) to(23) atmosphere,(24) recondens,(25) and(26) repeats.(27) The(28) cycle(29) of(30) birth,(31) flow,(30) and(31) death (30) sustizes(30) life,(30) shaping(30) ecosystems,(30) nourishing(30) bodies,(30) and(30) sustaining(30) the(30) planet.(30)
We have more than 30 words. Need exactly 50 words. Let's craft 50 words: "Water begins as vapor in the atmosphere, condens into droplets, travels through rivers, oceans, and clouds, evaporates back into air, recondens, and repeats. This cycle of birth, flow, and death sustures life, shapes ecosystems, nourizes bodies, and sust
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}
}