© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 71019
    Prev
    Next
    INDEX
    Explanations

    actions followed by prepositions

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

    The neuron responds to inflected action words—especially verbs in past‐participle (“-ed”) or gerund (“-ing”) form.

    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
     and
    -1.53
     cremallera
    -1.08
     которые
    -1.06
     iArr
    -1.02
     sillas
    -1.01
     БЕ
    -1.00
     التي
    -0.99
     toallas
    -0.98
     الكبير
    -0.98
     informace
    -0.96
    POSITIVE LOGITS
     owing
    1.14
     something
    1.09
     there
    1.09
     someone
    1.06
     in
    1.04
     into
    1.03
     fep
    1.01
     being
    1.00
     merito
    0.99
    њу
    0.98
    Activations Density 0.069%

    No Known Activations