© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128387
    Prev
    Next
    INDEX
    Explanations

    This neuron primarily detects apostrophe characters (single-quote marks), especially as used in contractions or quotes.

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

    foreign words and phrases

    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
     gallina
    -1.27
     angelo
    -1.20
    评估
    -1.16
    ｾ
    -1.16
    gemaakt
    -1.15
     заре
    -1.12
     leone
    -1.11
    itel
    -1.10
     pue
    -1.09
     talle
    -1.09
    POSITIVE LOGITS
     this
    1.52
    '
    1.34
     възможно
    1.27
     desemple
    1.27
     този
    1.24
     fanden
    1.18
    readFileSync
    1.16
    ↵↵
    1.13
    uties
    1.13
    了她
    1.13
    Activations Density 0.005%

    No Known Activations