© 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. 71856
    Prev
    Next
    INDEX
    Explanations

    lists of examples

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

    The neuron responds to colon characters (":"), i.e. it detects colon punctuation 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
    それにしても
    -1.45
    ちなみに
    -1.35
    があった
    -1.29
    というわけで
    -1.27
     was
    -1.23
    そのため
    -1.21
    ımda
    -1.21
    Viitteet
    -1.20
     включая
    -1.19
    を設置
    -1.19
    POSITIVE LOGITS
    になったら
    1.04
    corrhi
    1.03
     Mannschaft
    1.01
     ‘
    1.00
     selber
    0.98
     modelleri
    0.97
    ↵
    0.97
    .
    0.97
    ,
    0.94
    ときの
    0.94
    Activations Density 0.053%

    No Known Activations