© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 91636
    Prev
    Next
    INDEX
    Explanations

    @ followed by code/citation identifiers

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

    The neuron is triggered by the “@” symbol (e.g. in annotations, citations, or compile directives).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     at
    -1.34
     especially
    -1.09
     Suddenly
    -1.00
     Using
    -0.96
     Usually
    -0.94
     Actually
    -0.94
     Perhaps
    -0.93
     AND
    -0.92
     by
    -0.90
     And
    -0.90
    POSITIVE LOGITS
    ⁨
    1.37
    #$
    1.23
    RISK
    1.20
    ਗ਼
    1.17
     cubre
    1.12
    providedIn
    1.06
    cuela
    1.05
     ‐
    1.03
     NINE
    1.03
    eleven
    1.02
    Activations Density 0.090%

    No Known Activations