© 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. 22-GEMMASCOPE-RES-131K
    7. 34448
    Prev
    Next
    INDEX
    Explanations

    code variable declarations and errors

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

    The neuron fires on code-definition tokens—especially declaration keywords like “var” and “const” and the uppercase identifier names used in constant/field definitions.

    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
     belangrij
    -1.12
     diferite
    -1.09
     decide
    -1.08
     where
    -1.01
     determine
    -0.98
    為主
    -0.95
     decided
    -0.94
     میں
    -0.93
     două
    -0.93
     determines
    -0.90
    POSITIVE LOGITS
     koko
    1.13
    勫
    1.09
     ویتامین
    1.07
     sakin
    1.05
     veren
    1.05
     offerte
    1.04
    ````
    1.03
     our
    1.02
     garan
    1.01
     permanentes
    1.01
    Activations Density 0.026%

    No Known Activations