Gemma 2-2B
PulseAugur coverage of Gemma 2-2B — every cluster mentioning Gemma 2-2B across labs, papers, and developer communities, ranked by signal.
- used by Sparse Autoencoders 90%
- used by DagsHub 90%
- instance of Gemma 2 9B 90%
- used by alphaXiv 70%
- used by CatalyzeX 70%
- used by ScienceCast 70%
- affiliated with Llama 3.2:1b 70%
- used by Gotit.pub 70%
- used by arXiv 70%
- instance of CatalyzeX 70%
- developed Sparse Autoencoders 70%
- authored by arXiv 50%
5 day(s) with sentiment data
-
New LLMs Developed for Moroccan Arabic and Arabizi Dialects
A developer has fine-tuned two open-source large language models, SILMA-9B-Darija and SILMA-2B-Darija, to better understand and generate Moroccan Arabic (Darija) and its informal Arabizi script. These models were traine…
-
New framework offers formal guarantees for LLM interpretability
A new formal verification framework has been developed to address the fragility of mechanistic interpretability in large language models. Researchers demonstrated that minor input changes can drastically alter the inter…
-
New AI alignment methods improve efficiency and multi-dimensional control · 3 sources tracked
Researchers are developing new methods for aligning AI models with human preferences, aiming to improve efficiency and performance. One approach, DSPA, uses inference-time steering to condition alignment on prompts, sho…
-
Small Qwen3 LLM on Old Phone Controls Desktop Browser
A demonstration showcases the Qwen3-0.6B language model, running on a 2017 Samsung Note 8, successfully controlling a desktop Google Chrome browser. The model processed structured page representations to perform tasks l…
-
Cross-model KV cache sharing promises to speed up multi-model AI inference
Two research papers propose a method called cross-model KV cache sharing to improve the efficiency of multi-model AI inference pipelines. This technique allows the key-value states computed by one model during its initi…
-
New 'patterning' technique debiases AI reward models, shows cross-model transfer
Researchers have developed a new technique called "patterning" to debias reward models used in AI training. This method reweights preference pairs based on their impact on benchmark losses, effectively reducing stylisti…
-
Hugging Face unveils efficient multimodal encoder NeoMME, study favors encoders for Indic NER
Hugging Face has introduced NeoMME, a new family of multilingual multimodal encoders designed for efficiency. Unlike many generative models, NeoMME uses a single bidirectional Transformer to process both text and image …
-
New method enables cross-model KV state sharing for LLMs
Researchers have developed a novel "universal context-reuse layer" that enables KV (key-value) state sharing between different large language models, even those with varying architectures, tokenizers, and scales. This c…
-
New research explores grammar's geometry in Transformer layers
A new research paper explores the geometric properties of language representations within Transformer models. The study investigates how the intrinsic dimensionality (ID) of these representations changes across layers a…
-
LLMs fine-tuned for malaria drug discovery outperform proprietary models
A new study introduces Malaria-Instruct, a dataset designed for malaria drug discovery using large language models (LLMs). The research evaluated several open-source LLMs, finding that fine-tuned models significantly ou…
-
LLMs exhibit congruency effects similar to human cognition in conflict tasks
Researchers have developed a novel verbal conflict task to investigate congruency effects in large language models, drawing parallels to psychological and neuroscience studies. The task involves prompts that elicit a de…
-
New method uses Koopman operator for model interpretability
Researchers have developed a new method for mechanistic interpretability called "Intrinsic Structure" that uses the Koopman operator to analyze the spectral properties of a model's internal dynamics. This approach aims …
-
HyperSAE uses Poincaré geometry to boost Sparse Autoencoder performance
A new PyTorch library called HyperSAE has been developed to improve the efficiency of Sparse Autoencoders (SAEs) by employing Poincaré hyperbolic geometry. This approach addresses the limitations of standard SAEs, which…
-
New LLM fine-tuning method targets performance and carbon emission break-even
Researchers have developed a new fine-tuning method that incorporates a differentiable energy surrogate to optimize for both performance and carbon emissions in Large Language Models (LLMs). This approach aims to achiev…
-
New Tiled SVD Method Extracts Network Mechanisms Directly From Weights
Researchers have developed a new method called column-tiled SVD to extract usable weight mechanisms directly from linear sites within neural networks. This approach identifies concepts within the network's weights thems…
-
New training method enhances LLM interpretability by reducing signal loss
Researchers have developed a new method called replacement-aware training to improve the interpretability of large language models. This technique trains sparse auto-encoders (SAEs) to be robust to errors introduced by …
-
Gemma 2-2B research finds active feature planes have less holonomy
A new research paper published on arXiv investigates the concentration of holonomy within specific feature planes of the Gemma 2-2B model. The study preregistered its methodology and analysis rules before inspecting the…
-
New 'prolepsis' phenomenon identified in small transformer models
Researchers have identified a phenomenon called 'prolepsis' in small transformer models, where the model commits to a decision early in its processing and cannot correct it. This commitment is sustained by task-specific…
-
Google's Gemma 2 models achieve high performance with efficient architecture
Google's new Gemma 2 models, particularly the 27B parameter version, are demonstrating significant performance gains through architectural innovations rather than just increased size. These models utilize a hybrid atten…
-
New research identifies actionable directions to mitigate AI model misalignment
Researchers have identified a method to detect and mitigate emergent misalignment in language models by analyzing activation directions. This approach, tested across four model families including Qwen2.5-1.5B, Gemma-2-2…