© 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. 84516
    Prev
    Next
    INDEX
    Explanations

    requires, must, experience, planned, released

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

    This neuron strongly responds to English content words in their inflected forms—especially verbs ending in “-ed” or “-s.”

    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
     andare
    -1.23
     these
    -1.20
    6
    -1.19
    These
    -1.16
     edilen
    -1.13
     餐廳
    -1.10
    érrez
    -1.09
     этого
    -1.09
    并非
    -1.09
     (),
    -1.08
    POSITIVE LOGITS
     btw
    1.20
     incidentally
    1.19
     przecież
    1.18
    boven
    1.14
     poetas
    1.13
    Œ
    1.12
     szint
    1.11
     itself
    1.11
    ̴
    1.09
    digde
    1.09
    Activations Density 0.133%

    No Known Activations