© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 1036
    Prev
    Next
    INDEX
    Explanations

    special keywords like static, dup, cond

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

    The neuron fires on numeric tokens—literal numbers, numeric identifiers (like “num,” “buf”, array indices in brackets), and bit‐shift constants—i.e. anything representing a numeric value or index.

    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
     Heilige
    -1.25
     kommunik
    -1.20
     čern
    -1.12
    fokus
    -1.10
     krás
    -1.10
    pompa
    -1.09
     маленький
    -1.09
    Alike
    -1.08
     höl
    -1.04
    ņi
    -1.01
    POSITIVE LOGITS
    viembre
    1.15
     Atl
    1.08
     ayat
    1.08
     Thessalon
    1.06
     Kec
    1.05
     Obat
    1.03
     Embal
    1.02
     Mousse
    1.02
     Ayam
    1.02
     Joaqu
    1.02
    Activations Density 0.007%

    No Known Activations