© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 42321
    Prev
    Next
    INDEX
    Explanations

    Jonas, CLR, Airbus, lotus

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on tokens that are written in all-caps or otherwise start with an uppercase letter—i.e. acronyms, model numbers, and proper names.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ⡪
    -2.36
    各样的
    -2.36
    ﻊ
    -2.34
    릭터
    -2.28
    able
    -2.25
    ating
    -2.22
    賅
    -2.13
    ίη
    -2.11
    𞤢
    -2.11
    dată
    -2.06
    POSITIVE LOGITS
            
    2.84
     ก็
    2.70
    al
    2.53
    "
    2.47
    }$
    2.42
    もの
    2.25
                                
    2.22
     mestre
    2.20
     signaled
    2.11
    </th>
    2.08
    Activations Density 0.006%

    No Known Activations