Downloads · 30 days
14
8% of all-time downloads
metehan777/cc-aeo-geo-smollm2-135m
cc-aeo-geo-smollm2-135m is a machine learning model from metehan777. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Full fine-tune of HuggingFaceTB/SmolLM2-135M-Instruct on 153,861 AEO/GEO/SEO examples from Common Crawl CC-MAIN-2026-21.
Downloads · 30 days
14
8% of all-time downloads
All-time downloads
178
Public
Parameters
135M
269 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors269 MB · 99%
From the Hugging Face model README
Full fine-tune of HuggingFaceTB/SmolLM2-135M-Instruct on 153,861 AEO/GEO/SEO examples from Common Crawl CC-MAIN-2026-21.
Trained on a single NVIDIA H100 (bf16, full fine-tune, Liger fused-linear-cross-entropy) in minutes.
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model = AutoModelForCausalLM.from_pretrained("metehan777/cc-aeo-geo-smollm2-135m", torch_dtype=torch.bfloat16, device_map="auto")
tok = AutoTokenizer.from_pretrained("metehan777/cc-aeo-geo-smollm2-135m")
messages = [
{"role": "system", "content": "You are an AEO/GEO/SEO expert."},
{"role": "user", "content": "What are the key pillars of Answer Engine Optimization?"},
]
inputs = tok.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True).to(model.device)
print(tok.decode(model.generate(inputs, max_new_tokens=400)[0][inputs.shape[-1]:], skip_special_tokens=True))
The model emits <tool_call>{"name": "brightdata_search", "arguments": {"query": "..."}}</tool_call>.
Run the search, feed back <tool_response>...</tool_response>, get a grounded answer. See tool_config.json.
| Base | HuggingFaceTB/SmolLM2-135M-Instruct |
| Dataset | metehan777/cc-aeo-geo-fulltext-CC-MAIN-2026-21 |
| Examples | 153,861 |
| Method | Full fine-tune, bf16, Liger fused kernels |
| Hardware | 1× H100 80GB |
| Epochs | 1 |