© 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. 594
    Prev
    Next
    INDEX
    Explanations

    according to

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

    The neuron fires on the cue “according to” (i.e. attribution or citation phrases introducing a source).

    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
    jamento
    -1.13
     efect
    -1.09
    èses
    -1.09
     discours
    -1.08
     ades
    -1.07
    钇
    -1.06
     protocole
    -1.05
    OPPO
    -1.05
    游戏亮点
    -1.05
    ceso
    -1.05
    POSITIVE LOGITS
     many
    1.70
     our
    1.70
     they
    1.42
     most
    1.41
     some
    1.41
     their
    1.34
     everyone
    1.29
     experts
    1.23
     my
    1.20
     according
    1.20
    Activations Density 0.024%

    No Known Activations