You can deploy this model and start earning money today!

AmanPriyanshu/
gpt-oss-13.1b-specialized-health_or_medicine-pruned-moe-only-19-experts

Name: gpt-oss-13.1b-specialized-health_or_medicine-pruned-moe-only-19-experts
Availability: InStock

Views

No views yet

Health Or Medicine GPT-OSS Model (19 Experts)

Project: https://amanpriyanshu.github.io/GPT-OSS-MoE-ExpertFingerprinting/

👥 Follow the Authors

Aman Priyanshu

Supriti Vijay

Introduction

This is a pruned variant of OpenAI's GPT-OSS-20B model, reduced to 19 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.

Model Architecture & Statistics

Metric	Value
Base Model	openai/gpt-oss-20b
Architecture	Mixture-of-Experts Transformer
Total Parameters	~13.1B (pruned from 21B)
Original Experts per Layer	32
Pruned Experts per Layer	19
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

Pruning Methodology

What is Expert Pruning?

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:

Analyzing Usage Patterns: Tracking which experts activate most frequently for specific tasks
Removing Underutilized Experts: Discarding experts with low activation rates for the target domain
Preserving Router Functionality: Maintaining the routing mechanism with fewer available experts

Our Approach

Data-Driven Selection: Used activation patterns from health or medicine evaluation tasks
Systematic Reduction: Reduced from 32 to 19 experts per layer
No Retraining: Direct removal without additional training steps

Performance & Applications

Pruning Benefits

Smaller Memory Footprint: 59.4% of original expert parameters
Reduced Computational Load: Fewer routing decisions during inference
Focused Capabilities: Retains experts relevant to health or medicine tasks

Use Cases

Speculative Decoding: Draft model for full GPT-OSS-20B
Resource-Constrained Deployment: Edge devices, mobile applications
Research: Study expert specialization in MoE models
Fine-tuning: Smaller base model for domain adaptation

Note: Performance may vary depending on how well the pruned experts match your specific use case.

Motivation & Expert Selection

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:

GPQA: Graduate-level questions in physics, chemistry, biology (Diamond & Expert subsets)
MMLU/MMLU-Pro: Comprehensive knowledge across 57+ subjects including science, medicine, law
SORRY-Bench: Safety evaluation across harmful content categories
Tulu3: Persona-driven instruction following with verifiable constraints
Polyglot-or-Not: Multilingual factual completion tasks

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 19 experts per layer.

Dataset & Analysis Foundation

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.

Pruning Methodology

Our approach involves:

Activation Analysis: Comprehensive evaluation of expert usage patterns across domain-specific tasks
Expert Ranking: Identification of the most frequently activated experts for target domains
Systematic Pruning: Reduction from 32 to 19 experts while preserving router functionality
Quality Validation: Testing to ensure maintained performance on target tasks

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.

Usage

CPU Inference

python

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# Load the specialized model on CPU
model = AutoModelForCausalLM.from_pretrained(
    "AmanPriyanshu/gpt-oss-13.1b-specialized-health_or_medicine-pruned-moe-only-19-experts", 
    torch_dtype=torch.bfloat16, 
    device_map="cpu", 
    trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-13.1b-specialized-health_or_medicine-pruned-moe-only-19-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)

Apple Silicon (MPS) Inference

python

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-13.1b-specialized-health_or_medicine-pruned-moe-only-19-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-13.1b-specialized-health_or_medicine-pruned-moe-only-19-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)

GPU Inference

For CUDA GPU inference, the models work with minimal modifications from the CPU example above. Simply change:

python

device_map="auto"  # Will automatically use GPU if available
torch_dtype=torch.bfloat16  # or torch.float16

Model Examples

The following examples demonstrate the model's performance on health or medicine tasks:

Example 1

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 performs three core functions that keep the body alive and functioning:

| Function | What it does | How it works |
|----------|--------------|--------------|
| **Pumping (circulation)** | Moves blood throughout the body, delivering oxygen, nutrients, hormones, and removing waste. | The left ventricle ejects oxygen‑rich blood into the aorta; the right ventricle sends oxygen‑poor blood into the pulmonary artery. |
| **Pressure regulation** | Maintains blood pressure within a range that allows tissues to receive adequate blood flow. | The heart’s contractility, heart rate, and the resistance of the vascular system together determine systolic and diastolic pressures. |
| **Electrical coordination** | Generates and propagates the electrical impulses that trigger each heartbeat, ensuring the chambers contract in a synchronized rhythm. | The sinoatrial (SA) node initiates the impulse, which travels through the atrioventricular (AV) node, bundle branches, and Purkinje fibers to the ventricles. |

Together, these functions keep the circulatory system operating efficiently, supporting all metabolic processes in the body.<|return|>

Example 2

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 |
|--------|----------|--------|
| **Basic nature** | Living, self‑replicating cells | Non‑living particles that need a host cell to reproduce |
| **Size** | ~0.2–2 µm (micrometers) | ~20–200 nm (nanometers) |
| **Structure** | Cell wall, membrane, cytoplasm, DNA (often circular), ribosomes, organelles (e.g., mitochondria in some) | Protein coat (capsid) surrounding genetic material (DNA or RNA); no cellular machinery |
| **Genetic material** | DNA (usually circular plasmids + chromosome) | DNA or RNA (single or double strand) |
| **Replication** | Independent: divide by mitosis (bacterial cell division) | Dependent: hijack host cell’s replication machinery |
| **Metabolism** | Can photosynthesize, oxidize, ferment, etc. | No metabolism; rely entirely on host cell |
| **Infection mechanism** | Can colonize surfaces, form biofilms, or invade tissues; often spread by contact, airborne droplets, or vectors | Attach to specific host receptors, penetrate, replicate inside host cell |
| **Treatment** | Antibiotics (target cell wall, protein synthesis, DNA replication, etc.) | Antiviral drugs (target viral enzymes or entry), vaccines; antibiotics ineffective |
| **Evolutionary classification** | Organisms (prokaryotes) | Pathogens (not classified as organisms) |
| **Examples** | *Escherichia coli*, *Staphylococcus aureus*, *Bacillus subtilis* | Influenza virus, HIV, SARS‑CoV‑2, HPV |

**Bottom‑line:**  
- **Bacteria** are true living cells that can survive and grow on their own.  
- **Viruses** are essentially “genetic packages” that need a living cell to make copies of themselves.<|return|>

Example 3

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 chronic metabolic disorder in which the body cannot use glucose (sugar) properly.  The two main types—Type 1 and Type 2—have overlapping symptoms but differ in their underlying causes.  Below is a concise overview of the most common signs, symptoms, and the key factors that lead to each type.

---

## 1.  Symptoms (Signs that may indicate diabetes)

| Symptom | Typical Onset | Why It Happens |
|---------|---------------|----------------|
| **Polyphagia** (excessive hunger) | Often early | Body can’t use glucose → cells starve → hunger signals |
| **Polydipsia** (excessive thirst) | Early | High blood sugar pulls water out of cells → dehydration |
| **Polyuria** (frequent urination) | Early | Kidneys try to excrete excess glucose → osmotic diuresis |
| **Weight loss** (especially in Type 1) | Early | Body burns fat & muscle for energy |
| **Fatigue / weakness** | Early | Cells lack glucose → low energy |
| **Blurred vision** | Early‑mid | Hyperglycemia swells the lens, altering focus |
| **Slow wound healing** | Mid‑to‑late | Poor circulation & immune function |
| **Frequent infections** (skin, urinary, thrush) | Mid‑to‑late | High glucose supports bacterial growth; immune dysfunction |
| **Peripheral neuropathy** (tingling, numbness) | Late | Chronic hyperglycemia damages nerves |
| **Foot ulcers / amputations** | Late | Neuropathy + poor circulation |
| **Diabetic ketoacidosis** (Type 1) | Sudden | Insulin deficiency → fat breakdown → ketones → acidosis |
| **Hyperosmolar hyperglycemic syndrome** (Type 2) | Sudden | Extremely high glucose → dehydration, altered mental status |

> **Note:** Many people, especially with Type 2, may have *no symptoms* for years. Routine screening is essential if risk factors are present.

---

## 2.  Causes (Underlying mechanisms)

### Type 1 Diabetes (T1D)

| Cause | Key Features |
|-------|--------------|
| **Autoimmune destruction of β‑cells** | The immune system attacks insulin‑producing cells in the pancreas. |
| **Genetic predisposition** | Certain H

Citation

If you use this model in your research, please cite:

bibtex

@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}
}

References & Resources

Original Model: OpenAI GPT-OSS Model Card
Model Hub: GPT-OSS-20B on Hugging Face
Expert Analysis Dataset: GPT-OSS-20B MoE Expert Activations
Project Page: GPT-OSS MoE Expert Fingerprinting
GitHub Repository: OpenAI GPT-OSS