Downloads · 30 days
12
9% of all-time downloads
AmanPriyanshu/gpt-oss-17.9b-specialized-health_or_medicine-pruned-moe-only-27-experts
gpt-oss-17.9b-specialized-health_or_medicine-pruned-moe-only-27-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
12
9% of all-time downloads
All-time downloads
133
Public
Parameters
17.9B
35.9 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors35.9 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 27 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 health or medicine 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 | ~17.9B (pruned from 21B) |
| Original Experts per Layer | 32 |
| Pruned Experts per Layer | 27 |
| 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 | Health Or Medicine |
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 medical domain model incorporates experts that activated highly during health and medical tasks from MMLU medical subjects. These experts specialize in clinical knowledge, anatomy, medical procedures, and health-related reasoning.
The expert selection process utilized our comprehensive analysis of router activation patterns across multiple evaluation benchmarks:
By identifying experts that consistently activated for health or medicine tasks, we created this specialized model that maintains domain expertise while significantly reducing computational requirements from 32 to 27 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-17.9b-specialized-health_or_medicine-pruned-moe-only-27-experts",
torch_dtype=torch.bfloat16,
device_map="cpu",
trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-17.9b-specialized-health_or_medicine-pruned-moe-only-27-experts")
# Generate with the model
messages = [
{"role": "user", "content": "What are the main functions of the human heart?"}
]
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-17.9b-specialized-health_or_medicine-pruned-moe-only-27-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-17.9b-specialized-health_or_medicine-pruned-moe-only-27-experts")
# Generate with the model
messages = [
{"role": "user", "content": "What are the main functions of the human heart?"}
]
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 health or medicine tasks:
Question: What are the main functions of the human heart?
Response:
<|channel|>analysis<|message|>Need concise answer.<|end|><|start|>assistant<|channel|>final<|message|>The human heart is a muscular pump that performs three core functions:
| Function | What it does | Why it matters |
|----------|--------------|----------------|
| **Pumping blood** | Contracts and relaxes rhythmically to move blood through the circulatory system. | Delivers oxygen, nutrients, hormones, and removes waste products to every cell. |
| **Maintaining blood pressure** | Generates enough force to keep blood moving against resistance in arteries and veins. | Keeps tissues perfused and allows the body to respond to changes in activity or posture. |
| **Regulating blood flow** | Adjusts rate and force of contraction (cardiac output) and controls valve opening/closing. | Matches oxygen delivery to metabolic demand (e.g., during exercise vs rest). |
Together, these functions keep the body’s cells supplied with the substances they need and remove metabolic waste, sustaining life.<|return|>
Question: Explain the difference between bacteria and viruses.
Response:
<|channel|>analysis<|message|>Need concise explanation.<|end|><|start|>assistant<|channel|>final<|message|>**Bacteria vs. Viruses – the key differences**
| Feature | Bacteria | Viruses |
|--------|----------|---------|
| **Cellular status** | True cells – prokaryotic, with a cell membrane, cytoplasm, ribosomes, and a cell wall (often peptidoglycan). | Non‑cellular particles – no membrane, no cytoplasm, no ribosomes. |
| **Size** | 0.5–5 µm (visible under light microscope). | 20–300 nm (visible only with electron microscopy). |
| **Genetic material** | DNA or RNA, usually a single circular chromosome; some have plasmids. | DNA or RNA, single or double stranded, but never both. |
| **Reproduction** | Binary fission (autonomous, self‑replicating). | Must hijack a host cell’s machinery to replicate. |
| **Metabolism** | Can be autotrophic or heterotrophic; many can generate energy (e.g., via respiration, photosynthesis). | No metabolism of their own; rely entirely on host cell enzymes. |
| **Structure** | Cell wall, cytoplasmic membrane, ribosomes, sometimes flagella or pili. | Protein coat (capsid) surrounding the genome; some have an outer lipid envelope. |
| **Life cycle** | Can grow, divide, and form colonies. | Infect a host cell, replicate, then exit (lysis or budding). |
| **Treatment** | Antibiotics (target cell wall, protein synthesis, DNA replication, etc.). | Antivirals (target viral enzymes or entry/exit steps); antibiotics ineffective. |
| **Disease examples** | *Streptococcus pneumoniae* (pneumonia), *Escherichia coli* (UTI). | Influenza virus, HIV, SARS‑CoV‑2. |
| **Transmission** | Often via contact, food, water, vectors. | Often via aerosols, bodily fluids, or contaminated surfaces. |
**Bottom line:**
Bacteria are living, self‑sufficient cells that can grow and reproduce on their own. Viruses are obligate parasites that need a host cell to replicate; they are essentially genetic material wrapped in a protein coat and have no metabolic machinery of their own.<|return|>
Question: What are the symptoms and causes of diabetes?
Response:
<|channel|>analysis<|message|>Need to answer.<|end|><|start|>assistant<|channel|>final<|message|>**Diabetes mellitus** is a group of metabolic disorders characterized by chronic hyperglycemia (high blood sugar). The two most common types—Type 1 and Type 2—have distinct causes, but many symptoms overlap because they all stem from elevated glucose levels and impaired insulin action or production.
---
## 1. Symptoms
| Symptom | Why it Happens | Typical Onset |
|---------|----------------|---------------|
| **Polyuria** (frequent urination) | Kidneys excrete excess glucose, pulling water with it. | Often early, especially in Type 1. |
| **Polydipsia** (excessive thirst) | Dehydration from fluid loss via urine. | Early. |
| **Polyphagia** (increased hunger) | Cells can’t use glucose, so the body signals for more food. | Early. |
| **Weight loss** (especially in Type 1) | Body breaks down fat and muscle for energy. | Early in Type 1; later in Type 2. |
| **Fatigue / Weakness** | Energy production is impaired. | Early. |
| **Blurred vision** | Fluid shifts in the lens of the eye. | Early. |
| **Slow‑healing cuts or infections** | High glucose impairs immune function and wound healing. | Variable. |
| **Numbness or tingling in hands/feet** (neuropathy) | Chronic high glucose damages nerves. | Usually later. |
| **Darkened skin patches** (acanthosis nigricans) | Often seen in insulin resistance. | Later, especially in Type 2. |
| **Recurrent infections** (UTIs, yeast, skin) | Immune suppression. | Variable. |
| **Unexplained weight loss** (Type 2) | Less common but can occur in advanced disease. | Late. |
> **Note:** Some people, especially those with Type 2 diabetes, may have *no symptoms* for years. Routine screening is important if you have risk factors.
---
## 2. Causes
### A. Type 1 Diabetes (T1D)
| Factor | Explanation |
|--------|-------------|
| **Autoimmune destruction of β‑cells** | The immune system mistakenly attacks insulin‑producing cells in the pancreas. |
| **Genetic predisposition** | Certain HLA genes increase risk. |
| **Environmental triggers** |
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}
}