© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 59241
    Prev
    Next
    INDEX
    Explanations

    reading properties of undefined

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

    The neuron fires on programming‐language code tokens—things like imports/exports, declarations (`declare`, `interface`), API names, identifiers, and other non-natural-language syntax.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    uangan
    -1.26
     vaikka
    -1.17
    ffl
    -1.13
     снего
    -1.11
    甚至
    -1.10
     dacă
    -1.06
    Lietu
    -1.05
     我们
    -1.04
     ſi
    -1.03
     situada
    -1.03
    POSITIVE LOGITS
     six
    1.23
    推荐阅读
    1.16
    raised
    1.16
    quartered
    1.12
    ulant
    1.11
     four
    1.09
    quely
    1.09
    providedIn
    1.08
    rageous
    1.07
     ample
    1.06
    Activations Density 0.142%

    No Known Activations