© 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. 26046
    Prev
    Next
    INDEX
    Explanations

    file paths from programming projects

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

    The neuron responds to numeric tokens (especially version or build numbers) in code paths and error tracebacks.

    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
     vele
    -1.05
    tabak
    -1.05
     many
    -1.02
     or
    -1.02
     textil
    -1.00
    ֢
    -0.98
     varias
    -0.98
     existente
    -0.97
     animales
    -0.96
     hoje
    -0.95
    POSITIVE LOGITS
     Эти
    1.11
    手掌
    1.10
     prieten
    1.10
     Olivenöl
    1.08
    These
    1.08
     lucru
    1.06
     DEB
    1.05
     obicei
    1.03
     sogen
    1.02
    Mó
    1.02
    Activations Density 0.028%

    No Known Activations