Downloads · 30 days
230
24% of all-time downloads
sana0756/hipponet-gemma-398moe-omega2
hipponet-gemma-398moe-omega2 is a text generation model from sana0756. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
HippoNet Phase Omega-2 is an advanced Hierarchical Tree Mixture-of-Experts (Tree-MoE) architecture built on top of Google Gemma-4B.
Downloads · 30 days
230
24% of all-time downloads
All-time downloads
952
Public
Parameters
8.2B
380 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors361 GB · 92%
How the weights are stored.
BF164.6B · 56%
From the Hugging Face model README
HippoNet Phase Omega-2 is an advanced Hierarchical Tree Mixture-of-Experts (Tree-MoE) architecture built on top of Google Gemma-4B.
It injects 14 Domain Anchors and 28 Sub-shelves (totaling 392 specialized experts per MoE layer) into 21 transformer layers, activated via a zero-synchronization Turbo V2 Dense Weight Mask engine.
0.3878 - 0.4068 at step 2000!import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
REPO_ID = "sana0756/hipponet-gemma-398moe-omega2"
tokenizer = AutoTokenizer.from_pretrained(REPO_ID)
model = AutoModelForCausalLM.from_pretrained(
REPO_ID,
torch_dtype=torch.bfloat16,
device_map="auto"
)
prompt = """<start_of_turn>user
Aşağıdaki Python fonksiyonunu analiz et ve optimize edilmiş halini açıkla:
def fibonacci(n):
if n <= 1: return n
return fibonacci(n-1) + fibonacci(n-2)<end_of_turn>
<start_of_turn>model
"""
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512, temperature=0.7, top_p=0.9)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
Apache 2.0