sroecker (Steffen Röcker)

upvoted a collection 7 days ago

🇩🇪German SFT and DPO datasets

Collection

Datasets that can be used for LLM training with axolotl, trl or llama_factory. • 30 items • Updated May 27 • 8

upvoted an article 11 days ago

Article

Welcome Gemma 2 - Google's new open LLM

12 days ago

• 86

upvoted a collection 12 days ago

Gemma 2 Release

Collection

10 items • Updated 12 days ago • 114

upvoted a paper 13 days ago

The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale

Paper • 2406.17557 • Published 14 days ago • 73

upvoted an article 14 days ago

Article

Fine-tuning Florence-2 - Microsoft's Cutting-edge Vision Language Models

15 days ago

• 131

upvoted an article 15 days ago

Article

BM25 for Python: Achieving high performance while simplifying dependencies with BM25S⚡

By

•

about 4 hours ago

• 25

upvoted a paper 16 days ago

GenQA: Generating Millions of Instructions from a Handful of Prompts

Paper • 2406.10323 • Published 25 days ago • 5

upvoted a collection 16 days ago

Embedding Model Datasets

Collection

A curated subset of the datasets that work out of the box with Sentence Transformers: https://huggingface.co/datasets?other=sentence-transformers • 67 items • Updated 6 days ago • 46

upvoted a collection 17 days ago

4M Models

Collection

Multimodal models from https://4m.epfl.ch/ • 14 items • Updated 24 days ago • 29

upvoted 2 papers 17 days ago

GLiNER multi-task: Generalist Lightweight Model for Various Information Extraction Tasks

Paper • 2406.12925 • Published 25 days ago • 18

Instruction Pre-Training: Language Models are Supervised Multitask Learners

Paper • 2406.14491 • Published 19 days ago • 76

upvoted a collection 17 days ago

Instruction Pre-Training

Collection

8 items • Updated 18 days ago • 24

upvoted a collection 18 days ago

Hermes

Collection

Nous' Flagship LLM Series • 23 items • Updated 11 days ago • 95

upvoted a paper 19 days ago

Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models

Paper • 2404.18796 • Published Apr 29 • 67

upvoted 3 collections 20 days ago

upvoted a collection 21 days ago

BM25S Indices

Collection

https://github.com/xhluca/bm25s • 14 items • Updated 20 days ago • 7

upvoted a collection 22 days ago

DeepSeekCoder-V2

Collection

4 items • Updated 25 days ago • 57

upvoted a collection 24 days ago

Small LLMs

Collection

Collection of Fine Tuned Small LLMs • 13 items • Updated May 25 • 2

upvoted a collection 25 days ago

FP8 LLMs for vLLM

Collection

Accurate FP8 quantized models by Neural Magic, ready for use with vLLM! • 16 items • Updated about 17 hours ago • 18

upvoted a paper 25 days ago

Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing

Paper • 2406.08464 • Published 27 days ago • 48

upvoted an article 25 days ago

Article

Putting RL back in RLHF

27 days ago

• 53

upvoted a collection 27 days ago

codestral-text2cypher

Collection

codestral finetuned for text2cypher • 3 items • Updated 29 days ago • 2

upvoted 4 collections about 1 month ago

Local Function Calling Gems

Collection

These are the best function calling LLMs one can run on less than 64GB VRAM/Unified Memory. I use these on a M1 Max Macbook 64GB. • 6 items • Updated 8 days ago • 3

Qwen2

Collection

Qwen2 language models, including pretrained and instruction-tuned models of 5 sizes, including 0.5B, 1.5B, 7B, 57B-A14B, and 72B. • 29 items • Updated Jun 6 • 237

GLM-4

Collection

GLM-4 Open Models • 4 items • Updated Jun 5 • 84

DeTikZify

Collection

Synthesizing Graphics Programs for Scientific Figures and Sketches with TikZ • 9 items • Updated Jun 3 • 2

upvoted 4 articles about 1 month ago

Article

Uncensor any LLM with abliteration

By

•

26 days ago

• 230

Article

Releasing Common Corpus: the largest public domain dataset for training LLMs

By

•

Mar 20

• 12

Article

How to directly access 150k+ Hugging Face Datasets with DuckDB and query using GPT-4o

By

•

May 31

• 10

Article

⚗️ 🔥 Building High-Quality Datasets with distilabel and Prometheus 2

By

•

Jun 3

• 21

upvoted a collection about 1 month ago

sentence-transformers-from-synthetic-data

Collection

Example of using distilabel to generate synthetic triplets data for fine-tuning a Sentence Transformer model • 4 items • Updated 18 days ago • 20

upvoted an article about 1 month ago

Article

Training and Finetuning Embedding Models with Sentence Transformers v3

May 28

• 116

upvoted a paper about 1 month ago

Granite Code Models: A Family of Open Foundation Models for Code Intelligence

Paper • 2405.04324 • Published May 7 • 14

upvoted a collection about 1 month ago

🤖Phi-3

Collection

6 items • Updated 6 days ago • 2

upvoted an article about 1 month ago

Article

GPU Poor Savior: Revolutionizing Low-Bit Open Source LLMs and Cost-Effective Edge Computing

By

•

May 25

• 9

upvoted 2 collections about 1 month ago

DiscoLeo 8B: Llama3 for German

Collection

Continued Pretraining on Llama3 8B to improve German linguistic capabilities. A collection of base and fine-tuned models and variants. • 5 items • Updated May 25 • 14

DiscoLeo 8B quants

Collection

A collection of different quantizations of the DiscoLeo models. • 3 items • Updated May 25 • 3

upvoted a collection about 2 months ago

C4AI Aya 23

Collection

Aya 23 is an open weights research release of an instruction fine-tuned model with highly advanced multilingual capabilities. • 3 items • Updated May 23 • 40

upvoted an article about 2 months ago

Article

⚗️ 🧑🏼‍🌾 Let's grow some Domain Specific Datasets together

By

•

Apr 29

• 27

upvoted a collection about 2 months ago

C4AI Command R Plus

Collection

C4AI Command R+ is an open weights research release of a 104B billion parameter model with highly advanced capabilities. • 3 items • Updated May 23 • 23

upvoted an article about 2 months ago

Article

Let's talk about LLM evaluation

By

•

May 23

• 93

upvoted 4 collections about 2 months ago

Phi-3

Collection

Phi-3 family of small language and multi-modal models. Language models are available in short- and long-context lengths. • 22 items • Updated May 31 • 362

CommonCatalog

Collection

Common Catalog, a dataset with Creative Commons licensed images and machine-generated caption pairs • 8 items • Updated May 16 • 13

M2-BERT Embeddings

Collection

Models and Datasets for M2-BERT and LoCoV1 • 10 items • Updated May 22 • 2

Yi-1.5 (2024/05)

Collection

10 items • Updated May 20 • 84

upvoted a paper 2 months ago

What matters when building vision-language models?

Paper • 2405.02246 • Published May 3 • 91

upvoted 2 collections 2 months ago

Granite Code Models

Collection

A series of code models trained by IBM licensed under Apache 2.0 license. We release both the base pretrained and instruct models. • 20 items • Updated 10 days ago • 145

SPPO

Collection

Self-Play Preference Optimization • 10 items • Updated 10 days ago • 9

upvoted 3 articles 2 months ago

Article

Saving Memory Using Padding-Free Transformer Layers during Finetuning

By

•

28 days ago

• 8

Article

Introducing Idefics2: A Powerful 8B Vision-Language Model for the community

Apr 15

• 146

Article

🧑‍⚖️ "Replacing Judges with Juries" using distilabel

By

•

May 3

• 17

upvoted a collection 2 months ago

llama 3 self-align experiments

Collection

Replicating the pipeline for StarCoder-2 Instruct on Llama-3-8B with some tweaks https://huggingface.co/blog/sc2-instruct • 4 items • Updated May 9 • 6

upvoted 2 articles 2 months ago

Article

Improving Prompt Consistency with Structured Generations

Apr 30

• 49

Article

StarCoder2-Instruct: Fully Transparent and Permissive Self-Alignment for Code Generation

Apr 29

• 70

upvoted a collection 2 months ago

LLaVA-Phi-3-mini

Collection

4 items • Updated Apr 28 • 12

upvoted 2 articles 2 months ago

Article

Design choices for Vision Language Models in 2024

By

•

Apr 16

• 21

Article

Post-OCR-Correction: 1 billion words dataset of automated OCR correction by LLM

By

•

Apr 26

• 12

upvoted an article 3 months ago

Article

LLM Comparison/Test: Llama 3 Instruct 70B + 8B HF/GGUF/EXL2 (20 versions tested and compared!)

By

•

Apr 24

• 53

Steffen Röcker PRO

AI & ML interests

Organizations

sroecker's activity

Welcome Gemma 2 - Google's new open LLM

Fine-tuning Florence-2 - Microsoft's Cutting-edge Vision Language Models

BM25 for Python: Achieving high performance while simplifying dependencies with *BM25S*⚡

Putting RL back in RLHF

Uncensor any LLM with abliteration

Releasing Common Corpus: the largest public domain dataset for training LLMs

How to directly access 150k+ Hugging Face Datasets with DuckDB and query using GPT-4o

⚗️ 🔥 Building High-Quality Datasets with distilabel and Prometheus 2

Training and Finetuning Embedding Models with Sentence Transformers v3

GPU Poor Savior: Revolutionizing Low-Bit Open Source LLMs and Cost-Effective Edge Computing

⚗️ 🧑🏼‍🌾 Let's grow some Domain Specific Datasets together

Let's talk about LLM evaluation

Saving Memory Using Padding-Free Transformer Layers during Finetuning

Introducing Idefics2: A Powerful 8B Vision-Language Model for the community

🧑‍⚖️ "Replacing Judges with Juries" using distilabel

Improving Prompt Consistency with Structured Generations

StarCoder2-Instruct: Fully Transparent and Permissive Self-Alignment for Code Generation

Design choices for Vision Language Models in 2024

Post-OCR-Correction: 1 billion words dataset of automated OCR correction by LLM

LLM Comparison/Test: Llama 3 Instruct 70B + 8B HF/GGUF/EXL2 (20 versions tested and compared!)

BM25 for Python: Achieving high performance while simplifying dependencies with BM25S⚡