Akhil Theerthala PRO

Akhil-Theerthala's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

updated a Space about 17 hours ago

Running

2

2

Comic Genesis AI

🎨

Generate professional comic panels with AI

liked a dataset about 23 hours ago

HuggingFaceFW/finepdfs

Preview • Updated about 3 hours ago • 5.06k • 248

liked a Space 2 days ago

Running

2

2

Comic Genesis AI

🎨

Generate professional comic panels with AI

published a Space 2 days ago

Running

2

2

Comic Genesis AI

🎨

Generate professional comic panels with AI

updated a dataset 5 days ago

Akhil-Theerthala/WikiFinance

Viewer • Updated 5 days ago • 42k • 59

updated a dataset 6 days ago

Akhil-Theerthala/WikiFinance

Viewer • Updated 5 days ago • 42k • 59

published a dataset 6 days ago

Akhil-Theerthala/WikiFinance

Viewer • Updated 5 days ago • 42k • 59

liked a model 9 days ago

bartowski/Akhil-Theerthala_Kuvera-8B-v0.1.0-GGUF

Text Generation • 8B • Updated Jun 6 • 586 • 2

liked a dataset 12 days ago

multimodal-reasoning-lab/Zebra-CoT

Viewer • Updated Jul 26 • 160k • 16.3k • 49

liked a dataset 13 days ago

WaltonFuture/Multimodal-Cold-Start

Viewer • Updated Jul 24 • 51.5k • 440 • 10

reacted to codelion's post with 🔥 13 days ago

view post

Post

4921

I recently added a recipe in ellora to improve reasoning capabilities to Gemma-3-1B using self-supervised learning. Model now shows step-by-step thinking in <think> tags before answering.

Logic puzzle accuracy: 61% → 84%. 3 hours training on single GPU. 🧠

Used GRPO where model generates multiple responses and learns to prefer better reasoning. Works surprisingly well for making smaller models more transparent.

🔗 Colab: https://colab.research.google.com/github/codelion/ellora/blob/main/Ellora_Recipe_2_Reasoning_LoRA_with_Self-Rewarding_GRPO.ipynb

🤗 Model: codelion/gemma-3-1b-it-reasoning-grpo-lora

💻 Code: https://github.com/codelion/ellora

See translation

• 
• 1 reply

·

🔥

11

11

👀

2

2

🚀

1

1

❤️

1

1

+

Reply

liked a model 13 days ago

microsoft/VibeVoice-1.5B

Text-to-Speech • 3B • Updated 7 days ago • 237k • 1.56k

liked a dataset 17 days ago

LinguaCustodia/Wikipedia-Finance-Articles

Viewer • Updated Apr 24 • 106k • 25 • 2

liked a model 17 days ago

LinguaCustodia/ClassiFin

0.1B • Updated Mar 27 • 10 • 3

upvoted a collection 19 days ago

LLM Reasoning Papers

Collection

improve reasoning capabilities of LLMs • 45 items • Updated Feb 18 • 6

upvoted a paper 19 days ago

STaR: Bootstrapping Reasoning With Reasoning

Paper • 2203.14465 • Published Mar 28, 2022 • 9

replied to their post 26 days ago

view reply

Sure! Would love to have a chat!

updated a collection 27 days ago

Kuvera - PersonalFinance- V3

Collection

The third iteration of personal finance advice generation, with dataset generated from streamlined pipelines and models trained on these datasets. • 4 items • Updated 27 days ago • 1