Downloads · 30 days
8
7% of all-time downloads
AmanPriyanshu/gpt-oss-15.5b-specialized-instruction_following-pruned-moe-only-23-experts
gpt-oss-15.5b-specialized-instruction_following-pruned-moe-only-23-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
8
7% of all-time downloads
All-time downloads
114
Public
Parameters
15.5B
31.1 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors31.1 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 23 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 | ~15.5B (pruned from 21B) |
| Original Experts per Layer | 32 |
| Pruned Experts per Layer | 23 |
| 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 23 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-15.5b-specialized-instruction_following-pruned-moe-only-23-experts",
torch_dtype=torch.bfloat16,
device_map="cpu",
trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-15.5b-specialized-instruction_following-pruned-moe-only-23-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-15.5b-specialized-instruction_following-pruned-moe-only-23-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-15.5b-specialized-instruction_following-pruned-moe-only-23-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 formal email.<|end|><|start|>assistant<|channel|>final<|message|>**Subject:** Request for Meeting – Discussion of Research Project Progress
Dear Professor [Last Name],
I hope this message finds you well. I am writing to request a brief meeting with you to discuss the current status of my research project on [brief topic], as well as to seek your guidance on the next steps.
Could we meet at one of the following times next week?
- **Monday, Aug 18, 2025, 10:30 – 11:00 AM** (in your office or via Zoom)
- **Wednesday, Aug 20, 2025, 2:00 – 2:30 PM** (in your office or via Zoom)
- **Friday, Aug 22, 2025, 9:00 – 9:30 AM** (in your office or via Zoom)
If any of these slots are inconvenient, I would be happy to adjust to a time that better fits your schedule.
Thank you for your time and consideration. I look forward to our conversation.
Sincerely,
[Your Full Name]
[Your Program / Department]
[University Name]
[Email Address] | [Phone Number]<|return|>
Question: Create a numbered list of steps for making scrambled eggs, using exactly 6 steps.
Response:
<|channel|>analysis<|message|>Need 6 steps numbered.<|end|><|start|>assistant<|channel|>final<|message|>1. **Gather ingredients and tools** – Get 2–3 eggs, a pinch of salt, pepper, a splash of milk or water (optional), a non‑stick skillet, a spatula, and a small bowl.
2. **Whisk the eggs** – Crack the eggs into the bowl, add salt, pepper, and the splash of milk or water, then beat until the mixture is uniformly light and slightly frothy.
3. **Heat the skillet** – Place the skillet over medium‑low heat and add a small pat of butter or a drizzle of oil, letting it melt and coat the surface.
4. **Pour in the eggs** – Once the butter is foamy but not browned, pour the whisked eggs into the skillet, spreading them evenly.
5. **Cook and stir gently** – Let the eggs sit for a few seconds until the edges set, then use the spatula to gently stir from the edges toward the center, allowing the uncooked portions to flow underneath. Continue until the eggs are softly set but still slightly runny.
6. **Serve immediately** – Remove the skillet from heat, transfer the scrambled eggs to a plate, and enjoy hot, optionally garnishing with fresh herbs or a sprinkle of cheese.<|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 cycles through Earth via evaporation, condensation, precipitation, infiltration, runoff, and storage. Sun heats oceans, causing evaporation into atmosphere. Clouds form, then rain falls, replenishing rivers, lakes, and groundwater. Some infiltrates, recharging aquifers; excess runs off to seas, completing the loop. This cycle sustains life, climate, and ecosystems." Count words. Let's count: Water(1) cycles(2) through(3) Earth(4) via(5) evaporation,(6) condensation,(7) precipitation,(8) infiltration,(9) runoff,(10) and(11) storage.(12) Sun(13) heats(14) oceans,(15) causing(16) evaporation(17) into(18) atmosphere.(19) Clouds(20) form,(21) then(22) rain(23) falls,(24) replenishing(25) rivers,(26) lakes,(27) and(28) groundwater.(29) Some(30) infiltrates,(31) recharging(32) aquifers;(33) excess(34) runs(35) off(36) to(37) seas,(38) completing(39) the(40) loop.(41) This(42) cycle(43) sustains(44) life,(45) climate,(46) and(47) ecosystems.(48) Need 2 more words. Add "daily" and "balance". So add at end: daily balance. That adds 2 words. Let's append: "daily balance." Count: daily(49) balance.(50). Good.<|end|><|start|>assistant<|channel|>final<|message|>Water cycles through Earth via evaporation, condensation, precipitation, infiltration, runoff, and storage. Sun heats oceans, causing evaporation into atmosphere. Clouds form, then rain falls, replenishing rivers, lakes, and groundwater. Some infiltrates, recharging aquifers; excess runs off to seas, completing the loop. This cycle sustains life, climate, and ecosystems daily balance.<|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}
}