© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 43297
    Prev
    Next
    INDEX
    Explanations

    long linear genome

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

    nouns related to technical or academic concepts (like 'long', 'world', 'farther', 'submission', 'spread', 'aero', 'router', 'editor').

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires most strongly on relatively uncommon or low‐frequency words (i.e. rare tokens).

    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
     vediamo
    -1.59
     vestito
    -1.58
     utilizzare
    -1.51
     funziona
    -1.48
     meravigli
    -1.48
     attiv
    -1.47
     podéis
    -1.47
     trovo
    -1.45
    лтемелер
    -1.44
     orologio
    -1.44
    POSITIVE LOGITS
    :
    1.65
    _
    1.43
     "
    1.41
    ,
    1.38
     každé
    1.27
     всеми
    1.23
     quadruple
    1.22
    après
    1.21
     from
    1.20
     varje
    1.19
    Activations Density 0.159%

    No Known Activations