amenur (Aramis)

upvoted a paper 12 days ago

MOSEL: 950,000 Hours of Speech Data for Open-Source Speech Foundation Model Training on EU Languages

Paper • 2410.01036 • Published 15 days ago • 14

upvoted an article 20 days ago

Article

Llama can now see and run on your device - welcome Llama 3.2

22 days ago

• 151

upvoted a collection 28 days ago

Moshi v0.1 Release

Collection

MLX, Candle & PyTorch model checkpoints released as part of the Moshi release from Kyutai. Run inference via: https://github.com/kyutai-labs/moshi • 13 items • Updated 29 days ago • 207

upvoted an article 28 days ago

Article

Fine-tuning LLMs to 1.58bit: extreme quantization made easy

29 days ago

• 150

upvoted an article about 1 month ago

Article

Scaling robotics datasets with video encoding

Aug 27

• 33

upvoted 2 articles 3 months ago

Article

Fine-tune Llama 3.1 Ultra-Efficiently with Unsloth

By

•

Jul 29

• 224

Article

Memory-efficient Diffusion Transformers with Quanto and Diffusers

Jul 30

• 52

upvoted a paper 4 months ago

TextGrad: Automatic "Differentiation" via Text

Paper • 2406.07496 • Published Jun 11 • 26

upvoted 2 articles 4 months ago

Article

Extracting Concepts from LLMs: Anthropic’s recent discoveries 📖

By

•

Jun 20

• 26

Article

Uncensor any LLM with abliteration

By

•

Jun 13

• 345

upvoted 3 articles 5 months ago

Article

License to Call: Introducing Transformers Agents 2.0

May 13

• 114

Article

makeMoE: Implement a Sparse Mixture of Experts Language Model from Scratch

By

•

May 7

• 38

Article

SeeMoE: Implementing a MoE Vision Language Model from Scratch

By

•

Jun 23

• 33

upvoted an article 6 months ago

Article

seemore: Implement a Vision Language Model from Scratch

By

•

Jun 23

• 60

upvoted 2 papers 6 months ago

Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

Paper • 2404.14219 • Published Apr 22 • 251

Mixture-of-Depths: Dynamically allocating compute in transformer-based language models

Paper • 2404.02258 • Published Apr 2 • 104

upvoted an article 6 months ago

Article

Custom architectures with HuggingFace 🤗

By

•

Apr 22

• 21

upvoted a paper 6 months ago

Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?

Paper • 2404.03411 • Published Apr 4 • 8

upvoted 4 papers 7 months ago

Transformer-Lite: High-efficiency Deployment of Large Language Models on Mobile Phone GPUs

Paper • 2403.20041 • Published Mar 29 • 34

upvoted a collection 7 months ago

Transformers compatible Mamba

Collection

This release includes the `mamba` repositories compatible with the `transformers` library • 5 items • Updated Mar 6 • 35

upvoted a paper 8 months ago

The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits

Paper • 2402.17764 • Published Feb 27 • 596

Aramis

AI & ML interests

Organizations

amenur's activity

MOSEL: 950,000 Hours of Speech Data for Open-Source Speech Foundation Model Training on EU Languages

Llama can now see and run on your device - welcome Llama 3.2

Moshi v0.1 Release

Fine-tuning LLMs to 1.58bit: extreme quantization made easy

Scaling robotics datasets with video encoding

Fine-tune Llama 3.1 Ultra-Efficiently with Unsloth

Memory-efficient Diffusion Transformers with Quanto and Diffusers

TextGrad: Automatic "Differentiation" via Text

Extracting Concepts from LLMs: Anthropic’s recent discoveries 📖

Uncensor any LLM with abliteration

License to Call: Introducing Transformers Agents 2.0

makeMoE: Implement a Sparse Mixture of Experts Language Model from Scratch

SeeMoE: Implementing a MoE Vision Language Model from Scratch

seemore: Implement a Vision Language Model from Scratch

Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

Mixture-of-Depths: Dynamically allocating compute in transformer-based language models

Custom architectures with HuggingFace 🤗

Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?

The Unreasonable Ineffectiveness of the Deeper Layers

Gecko: Versatile Text Embeddings Distilled from Large Language Models

Jamba: A Hybrid Transformer-Mamba Language Model

Transformer-Lite: High-efficiency Deployment of Large Language Models on Mobile Phone GPUs

Transformers compatible Mamba

The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits