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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 71842
    Prev
    Next
    INDEX
    Explanations

    no matter what

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

    The neuron fires most strongly on relatively uncommon “content” tokens (i.e. low‐frequency, information‐bearing words) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     dueños
    -1.30
    部分が
    -1.29
     and
    -1.27
     agujeros
    -1.23
     astonishing
    -1.20
    marzo
    -1.20
    なのでしょう
    -1.19
     dragón
    -1.17
    这样
    -1.17
    化した
    -1.16
    POSITIVE LOGITS
    !
    2.41
    </h2>
    1.59
    ):
    1.59
    fordshire
    1.55
    );
    1.54
    oczes
    1.50
    。「
    1.34
    !)
    1.30
    })();
    
    1.30
    '];
    
    1.28
    Activations Density 0.084%

    No Known Activations