© 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. 127793
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on the generic placeholder word “things” (and its immediate context in vague “things that…” constructions).

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

    things that like

    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
     作り方
    -1.09
     Napoléon
    -1.05
    Reklama
    -1.03
    تحميل
    -1.02
    Применение
    -1.00
     Pilih
    -0.98
    ꒰
    -0.96
    Pancake
    -0.95
    wendi
    -0.94
    徳島
    -0.94
    POSITIVE LOGITS
     that
    1.69
     you
    1.48
     about
    1.36
     everyone
    1.30
     happened
    1.29
     happening
    1.20
     everything
    1.12
     happen
    1.09
     your
    1.05
     which
    1.04
    Activations Density 0.059%

    No Known Activations