lol thanks man
Hoglet (Ash) PRO
Hoglet-33
AI & ML interests
Open source AI, datasets, non-attention based models, parameter efficiency, SLMs, AI for the betterment of humanity. Contact at ash@basicallyai.co
Recent Activity
new activity 1 day ago
OpenModel-Labs/Comet-1-Preview:Here updated a Space 4 days ago
basically-ai/README new activity 4 days ago
AxiomicLabs/Open_SLM_Leaderboard:Weight EditingOrganizations
replied to their post 14 days ago
reacted to their post with 😔❤️👀🔥 15 days ago
Post
3,955
I wish Hugging Face posts had polls...
- 6 replies
posted an update 15 days ago
Post
3,955
I wish Hugging Face posts had polls...
- 6 replies
reacted to Banaxi-Tech's post with 😔 20 days ago
Post
202
So... im on vacation for 5 more days and I just realized... my hotel doesent have a grounding port 😟 so my pc shocked me 3 times.
So no progress for 5 days... only me and this phone
So no progress for 5 days... only me and this phone
- 2 replies
reacted to DedeProGames's post with 🚀🔥 26 days ago
Post
8,940
🚀 Introducing Kiyo-230M-Instruct — my compact English & Portuguese chat model!
Built on Kiyo-230M-Preview, a base model I pretrained from scratch, Kiyo now has an instruction-tuned release for everyday conversations, rewriting, and lightweight assistant experiments.
🧠 229.7M parameters
🌍 English + Portuguese
💬 Native chat template for multi-turn conversations
📏 2,048-token context window
📦 459 MB of BF16 safetensors weights
🔓 Apache-2.0 license
For instruction tuning, I used 165,735 conversations from SmolTalk2, smol-smoltalk, UltraChat, and Dolci-Instruct-SFT. Training ran for two complete epochs with assistant-only loss, using a roughly 70% English / 30% Portuguese token mix.
The updated model card includes benchmarks against SmolLM2-135M-Instruct, BananaMind, and Supra2, plus example conversations, CPU measurements, training details, and a Transformers quickstart.
The results are mixed: Kiyo is competitive with these small models across the reported evaluations, while SmolLM2 and BananaMind score higher on the instruction-following benchmarks. Portuguese was a highlight in the small conversational sanity test.
This is an experimental 230M model, with plenty of room to improve reasoning, factual accuracy, and strict instruction following. I’m excited to share it with people exploring how much we can get from small, accessible models!
🤗 Model & full README:
DedeProGames/Kiyo-230M-Instruct
💬 Try the chat demo:
DedeProGames/kiyo-230m-chat
Try it in English or Portuguese — I’d love to see your examples, feedback, and experiments!
Built on Kiyo-230M-Preview, a base model I pretrained from scratch, Kiyo now has an instruction-tuned release for everyday conversations, rewriting, and lightweight assistant experiments.
🧠 229.7M parameters
🌍 English + Portuguese
💬 Native chat template for multi-turn conversations
📏 2,048-token context window
📦 459 MB of BF16 safetensors weights
🔓 Apache-2.0 license
For instruction tuning, I used 165,735 conversations from SmolTalk2, smol-smoltalk, UltraChat, and Dolci-Instruct-SFT. Training ran for two complete epochs with assistant-only loss, using a roughly 70% English / 30% Portuguese token mix.
The updated model card includes benchmarks against SmolLM2-135M-Instruct, BananaMind, and Supra2, plus example conversations, CPU measurements, training details, and a Transformers quickstart.
The results are mixed: Kiyo is competitive with these small models across the reported evaluations, while SmolLM2 and BananaMind score higher on the instruction-following benchmarks. Portuguese was a highlight in the small conversational sanity test.
This is an experimental 230M model, with plenty of room to improve reasoning, factual accuracy, and strict instruction following. I’m excited to share it with people exploring how much we can get from small, accessible models!
🤗 Model & full README:
DedeProGames/Kiyo-230M-Instruct
💬 Try the chat demo:
DedeProGames/kiyo-230m-chat
Try it in English or Portuguese — I’d love to see your examples, feedback, and experiments!
- 9 replies
replied to DedeProGames's post 26 days ago
@DedeProGames Why those languages?
reacted to wayneworkman2012's post with 👀➕🚀 29 days ago
Post
3,199
To all the inference providers behind the HuggingFace router - if you want me to use your services you will have to serve native-precision models. I'm not paying for FP8 when I can run in bf16 locally.
- 6 replies
replied to Banaxi-Tech's post 29 days ago
@Banaxi-Tech YES! you should. You are the one that inspired me to start making SLMs in the first place. Just because of some controversy doesn't mean you should stop.
reacted to Banaxi-Tech's post with 😔❤️ 29 days ago
Post
4,235
Should I even continue making models?
- 27 replies
replied to their post about 1 month ago
yeah def, it might be easier to do one launch for these though... depends i guess.
replied to their post about 1 month ago
@Datdanboi25 just wondering why you would prefer the models to be launched when ready instead of in one big launch?
replied to Banaxi-Tech's post about 1 month ago
yeah i know, i have my priorities elsewhere
replied to Banaxi-Tech's post about 1 month ago
YAY!
also dang it, i was gonna do that lol. How do you build all of these so quickly?