© 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. 22-GEMMASCOPE-RES-131K
    7. 130312
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly responds to tokens that are part of product or exam identifiers—i.e. model/series names and codes combining letters and numbers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    professional identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    ISTAS
    -1.19
    melden
    -1.19
    RUPT
    -1.14
    ignal
    -1.13
     jaja
    -1.09
    moiselle
    -1.09
    ⌗
    -1.09
    ilets
    -1.08
    lojik
    -1.07
    termilk
    -1.06
    POSITIVE LOGITS
     you
    1.09
     noastre
    1.05
     offenbar
    1.05
     opportunità
    1.04
     acestei
    0.99
     tecnologie
    0.97
     stesse
    0.95
    却是
    0.94
     față
    0.94
     attività
    0.91
    Activations Density 0.001%

    No Known Activations