2 10 11

Pretam Ray

Pretam

raypretam

AI & ML interests

NLP

Recent Activity

upvoted a collection 9 days ago

Qwen3

upvoted a paper 18 days ago

GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models

upvoted a paper 19 days ago

Seed-Prover: Deep and Broad Reasoning for Automated Theorem Proving

View all activity

Organizations

upvoted a collection 9 days ago

Qwen3

Collection

84 items • Updated Aug 6 • 1.2k

upvoted a paper 18 days ago

GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models

Paper • 2508.06471 • Published about 1 month ago • 175

upvoted a paper 19 days ago

Seed-Prover: Deep and Broad Reasoning for Automated Theorem Proving

Paper • 2507.23726 • Published Jul 31 • 112

upvoted a collection about 1 month ago

DepNeCT

Collection

This Hugging Face collection hosts models and datasets from DepNeCT — a dependency-based method for nested compound type identification in Sanskrit • 4 items • Updated Jul 29 • 2

liked a model about 2 months ago

nvidia/OpenReasoning-Nemotron-32B

Text Generation • 33B • Updated 10 days ago • 2.54k • • 110

updated a model 2 months ago

Pretam/hindi_sanskrit

0.6B • Updated Jul 3 • 2

published a model 2 months ago

Pretam/hindi_sanskrit

0.6B • Updated Jul 3 • 2

authored a paper 4 months ago

REFINE-AF: A Task-Agnostic Framework to Align Language Models via Self-Generated Instructions using Reinforcement Learning from Automated Feedback

Paper • 2505.06548 • Published May 10 • 30

upvoted a paper 4 months ago

REFINE-AF: A Task-Agnostic Framework to Align Language Models via Self-Generated Instructions using Reinforcement Learning from Automated Feedback

Paper • 2505.06548 • Published May 10 • 30

liked a model 5 months ago

google/gemma-3-4b-it-qat-int4-unquantized

Image-Text-to-Text • 4B • Updated Apr 15 • 2.83k • 8

updated a model 5 months ago

Pretam/lora_model_gemma-3-12b-it_anushtup_final

Updated Mar 27

published a model 5 months ago

Pretam/lora_model_gemma-3-12b-it_anushtup_final

Updated Mar 27

liked 2 Spaces 6 months ago

584

Scaling test-time compute

📈

Implement test-time compute scaling for math problems

971

Model Memory Utility

🚀

Calculate vRAM needed for model training and inference

published a model 7 months ago

Pretam/t5-small-finetuned-xsum

Updated Jan 30

upvoted an article about 1 year ago

Article

Making LLMs even more accessible with bitsandbytes, 4-bit quantization and QLoRA

and 4 others •

May 24, 2023

• 163

reacted to vladbogo's post with 👍 over 1 year ago

Post

A recent paper titled "ShortGPT: Layers in Large Language Models are More Redundant Than You Expect" proposes a simple and effective approach to pruning Large Language Models (LLMs) by removing redundant layers.

Key points:
* Discovers significant redundancy across layers in LLMs, with some layers playing a negligible role for the final performance.
* Defines a new metric called Block Influence (BI) to quantify the importance of each layer in an LLM.
* Removes layers with low BI scores, achieving up to 25% reduction in parameters and computation while maintaining 92% of the LLM's performance.

Congrats to the authors for their work!

Paper: ShortGPT: Layers in Large Language Models are More Redundant Than You Expect (2403.03853)