© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 127741
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on auxiliary verbs that kick off a question (e.g. “Does…”, “Will…”)—i.e. it detects the start of an interrogative sentence.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    desired output or outcome

    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
    Negative Logits
    Spotify
    -0.85
     ones
    -0.84
    verze
    -0.82
    Orang
    -0.81
    尔夫
    -0.79
    Condu
    -0.78
     faço
    -0.78
     プレゼント
    -0.78
     *}$
    -0.78
    Assumptions
    -0.78
    POSITIVE LOGITS
     yrity
    1.00
     jät
    0.85
     financières
    0.82
     télévis
    0.82
     zależ
    0.81
     推薦
    0.81
     trygg
    0.81
     väli
    0.81
     kyse
    0.81
     undisclosed
    0.80
    Activations Density 0.238%

    No Known Activations