© 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. 95734
    Prev
    Next
    INDEX
    Explanations

    names followed by descriptions

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

    The neuron is sensitive to unusually long contiguous character sequences, i.e. it activates on very long 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_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
     butik
    -1.20
     Werken
    -1.20
     of
    -1.16
     legenda
    -1.13
    徘
    -1.08
     kanton
    -1.05
    BoxShadow
    -1.05
     foton
    -1.05
     menn
    -1.04
     perioden
    -1.04
    POSITIVE LOGITS
     deoarece
    1.14
     Auftritt
    1.13
     jeszcze
    1.11
     Rückseite
    1.11
     jähr
    1.11
     cauza
    1.07
     Kampagne
    1.06
    様な
    1.05
    ].
    1.05
    畢竟
    1.04
    Activations Density 0.000%

    No Known Activations