Downloads · 30 days
13
11% of all-time downloads
AmanPriyanshu/gpt-oss-16.7b-specialized-science-pruned-moe-only-25-experts
gpt-oss-16.7b-specialized-science-pruned-moe-only-25-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
13
11% of all-time downloads
All-time downloads
117
Public
Parameters
16.7B
33.5 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors33.5 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 25 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 science 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 | ~16.7B (pruned from 21B) |
| Original Experts per Layer | 32 |
| Pruned Experts per Layer | 25 |
| 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 | Science |
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 science-specialized model leverages experts that showed high activation patterns during scientific reasoning tasks from GPQA (physics, chemistry, biology) and MMLU science domains. These experts demonstrate superior performance on complex scientific problem-solving and technical knowledge recall.
The expert selection process utilized our comprehensive analysis of router activation patterns across multiple evaluation benchmarks:
By identifying experts that consistently activated for science tasks, we created this specialized model that maintains domain expertise while significantly reducing computational requirements from 32 to 25 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-16.7b-specialized-science-pruned-moe-only-25-experts",
torch_dtype=torch.bfloat16,
device_map="cpu",
trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("AmanPriyanshu/gpt-oss-16.7b-specialized-science-pruned-moe-only-25-experts")
# Generate with the model
messages = [
{"role": "user", "content": "Explain the process of photosynthesis in plants."}
]
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-16.7b-specialized-science-pruned-moe-only-25-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-16.7b-specialized-science-pruned-moe-only-25-experts")
# Generate with the model
messages = [
{"role": "user", "content": "Explain the process of photosynthesis in plants."}
]
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 science tasks:
Question: Explain the process of photosynthesis in plants.
Response:
<|channel|>analysis<|message|>Need concise explanation.<|end|><|start|>assistant<|channel|>final<|message|>**Photosynthesis** is the process by which green plants, algae, and some bacteria convert light energy into chemical energy, producing sugars that fuel growth and releasing oxygen as a by‑product. It takes place mainly in the chloroplasts of plant cells, specifically within the thylakoid membranes.
---
## 1. Light‑Dependent Reactions (Photophosphorylation)
| Step | What Happens | Key Components |
|------|--------------|-----------------|
| **Photon absorption** | Light energy excites electrons in chlorophyll a and b. | Chlorophyll, accessory pigments |
| **Water splitting (photolysis)** | H₂O → ½ O₂ + 2 H⁺ + 2 e⁻ (electrons). | Photosystem II (PSII), oxygen‑evolving complex |
| **Electron transport chain (ETC)** | Excited electrons travel from PSII → plastoquinone → cytochrome b₆f → plastocyanin → Photosystem I (PSI). | PSII, PQ, Cyt b₆f, PC, PSI |
| **ATP synthesis** | Proton gradient (ΔpH) drives ATP synthase. | ATP synthase, thylakoid lumen |
| **NADPH formation** | Electrons reduce NADP⁺ to NADPH. | PSI, ferredoxin, NADP⁺ reductase |
| **O₂ release** | Oxygen produced from water splitting is released to the atmosphere. | PSII, oxygen-evolving complex |
**Outcome of light‑dependent reactions:**
- **2 ATP** (per NADPH produced)
- **2 NADPH**
- **½ O₂** (per 2 H₂O)
---
## 2. Light‑Independent Reactions (Calvin Cycle)
The Calvin cycle uses ATP and NADPH to fix atmospheric CO₂ into sugars. It occurs in the stroma of the chloroplast.
| Phase | Key Enzyme | Reaction |
|-------|------------|----------|
| **Carbon fixation** | Ribulose‑1,5‑bisphosphate carboxylase/oxygenase (RuBisCO) | CO₂ + RuBP → 2 3‑phosphoglycerate (3‑PGA) |
| **Reduction** | Glyceraldehyde‑3‑phosphate dehydrogenase (GAPDH) | 3‑PGA + ATP + NADPH
Question: What causes the greenhouse effect and how does it work?
Response:
<|channel|>analysis<|message|>Need explain greenhouse effect causes, mechanism, greenhouse gases, absorption, re-emission, radiation balance.<|end|><|start|>assistant<|channel|>final<|message|>### What Causes the Greenhouse Effect?
The **greenhouse effect** is the natural process that warms the Earth’s surface to a temperature that can support life. It is caused by certain gases in the atmosphere—called **greenhouse gases (GHGs)**—that absorb and re‑emit infrared (IR) radiation. The main contributors are:
| Greenhouse Gas | Natural Sources | Anthropogenic (human) Sources | Typical Atmospheric Concentration (ppmv) |
|-----------------|------------------|------------------------------|----------------------------------------|
| **Water vapor (H₂O)** | Evaporation from oceans, lakes, soil | None (indirectly increased by warming) | 400–6000 (variable) |
| **Carbon dioxide (CO₂)** | Respiration, volcanic eruptions | Fossil‑fuel combustion, deforestation, cement production | ~415 |
| **Methane (CH₄)** | Wetlands, termites, ruminants | Agriculture (livestock, rice paddies), landfills, oil & gas extraction | ~1.9 |
| **Nitrous oxide (N₂O)** | Soil microbes, decomposition | Agricultural fertilizers, combustion | ~0.32 |
| **Ozone (O₃)** | Photochemical reactions | Industrial emissions, vehicle exhaust | ~0.5 |
| **CFCs & other halocarbons** | None | Refrigeration, aerosol propellants, foam blowing | <0.01 |
> **Key point:** Water vapor is the most abundant GHG, but it is *feedback‑controlled* by temperature. CO₂, CH₄, N₂O, and the halocarbons are *forcing agents* that directly change the radiative balance.
---
### How Does the Greenhouse Effect Work?
1. **Solar Radiation (Short‑wave)**
- The Sun emits mostly visible and near‑infrared light (≈ 0.3–2.5 µm).
- The atmosphere is largely transparent to this short‑wave radiation; it reaches the Earth’s surface.
2. **Absorption & Heating of the Surface**
- The surface absorbs the incoming solar energy and warms.
- A warmer surface emits thermal radiation (long‑wave, ≈ 5–50 µm).
3. **Interaction with Greenhouse Gases**
- Greenhouse gases
Question: Describe the structure and function of DNA.
Response:
<|channel|>analysis<|message|>Need concise description.<|end|><|start|>assistant<|channel|>final<|message|>**DNA (Deoxyribonucleic Acid)** is the hereditary material that carries the genetic instructions used in the growth, development, functioning, and reproduction of all living organisms and many viruses. Its structure and function can be understood in two complementary ways: the *molecular architecture* that gives it its physical properties, and the *biological roles* it plays in life.
---
## 1. Molecular Structure
| Feature | Description | Biological Significance |
|---------|-------------|--------------------------|
| **Double‑helix** | Two complementary strands wound around each other in a right‑handed helix. | Provides stability and a compact, yet flexible, storage form. |
| **Backbone** | Alternating deoxyribose sugars and phosphate groups. | Forms a sugar‑phosphate backbone that is chemically robust and resistant to enzymatic degradation. |
| **Bases** | Four nitrogenous bases: Adenine (A), Thymine (T), Cytosine (C), Guanine (G). | Encode genetic information; base pairing (A‑T, C‑G) ensures accurate replication. |
| **Base pairing** | Hydrogen bonds: A–T (2 bonds), C–G (3 bonds). | Maintains the double‑helix geometry and allows precise copying of genetic information. |
| **Antiparallel orientation** | One strand runs 5’→3’, the other 3’→5’. | Essential for replication and transcription machinery to recognize directionality. |
| **Major & minor grooves** | Structural features that allow protein binding. | Facilitate interaction with enzymes, transcription factors, and other regulatory proteins. |
| **Chromatin packaging** | DNA wrapped around histone proteins to form nucleosomes → chromatin → chromosomes. | Enables compaction of long DNA strands into the nucleus and regulates gene accessibility. |
---
## 2. Biological Functions
| Function | How DNA accomplishes it | Key Players |
|----------|------------------------|-------------|
| **Genetic information storage** | Sequence of bases encodes proteins, RNAs, and regulatory elements. | Genes, regulatory sequences (promoters, enhancers). |
| **Replication** | Semi‑conservative copying: each new duplex contains one old and one new strand. | DNA polymerases, helicases, primases, ligases. |
| **Transcription** | DNA → messenger RNA (mRNA) via RNA polymerase
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}
}