© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14878
    Prev
    Next
    INDEX
    Explanations

    reveals information or states something

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

    This neuron activates on semantically content-rich words (longer nouns, verbs, and adjectives) that carry narrative meaning.

    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
     we
    -1.23
     you
    -1.16
     and
    -1.07
     here
    -1.02
    cloudf
    -0.99
     amikor
    -0.98
     você
    -0.97
     or
    -0.91
    }$,
    -0.90
     dô
    -0.90
    POSITIVE LOGITS
    entait
    1.38
    めていた
    1.30
    内容は
    1.13
    noinspection
    1.13
     LIABLE
    1.10
     نداشت
    1.09
     kafe
    1.09
    中には
    1.09
    めている
    1.05
    Koordinaten
    1.05
    Activations Density 0.019%

    No Known Activations