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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 350
    Prev
    Next
    INDEX
    Explanations

    hexadecimal identifiers

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

    The neuron activates on hexadecimal identifier strings (e.g. commit hashes or long hex IDs).

    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
     or
    -1.01
     lesz
    -1.00
    xffffffff
    -0.98
     Arkivert
    -0.94
     okt
    -0.92
     kompon
    -0.90
    🎄
    -0.89
     had
    -0.88
    \}.
    -0.88
     metode
    -0.87
    POSITIVE LOGITS
     面包
    1.31
     zostały
    1.20
    くれます
    1.13
    essentiel
    1.09
     capilar
    1.05
    CAPÍTULO
    1.03
     倉
    1.02
     Alkaline
    1.02
    也被
    1.01
     useAppContext
    1.01
    Activations Density 0.002%

    No Known Activations