© 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. 22-GEMMASCOPE-RES-131K
    7. 54386
    Prev
    Next
    INDEX
    Explanations

    answers to questions

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

    The neuron activates on relational function words (like “of,” “after,” “in,” “by,” “value”) that introduce or connect numbers in math‐question contexts.

    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
    IVING
    -1.43
    ená
    -1.41
     великолеп
    -1.41
    Descriere
    -1.40
    μένου
    -1.38
     naves
    -1.38
    跸
    -1.38
    Kelebihan
    -1.36
    푉
    -1.36
    дной
    -1.33
    POSITIVE LOGITS
    )
    1.72
    —
    1.66
    .
    1.65
    !
    1.59
    </b>
    1.53
    "
    1.43
    ;
    1.43
    ?
    1.41
    就可以了
    1.41
    <
    1.30
    Activations Density 0.103%

    No Known Activations