© 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. 39832
    Prev
    Next
    INDEX
    Explanations

    code blocks and definitions

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

    The neuron is specialized to pick out numeric literals in source code (i.e. number tokens).

    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
    providedIn
    -1.42
     lørdag
    -1.21
    着脸
    -1.14
    PENDAHULUAN
    -1.11
     zimowa
    -1.10
     nyel
    -1.09
     kabát
    -1.09
     veze
    -1.09
    roxene
    -1.09
     szeret
    -1.08
    POSITIVE LOGITS
    if
    1.45
     zuerst
    1.07
     if
    1.06
     ď
    1.06
     firstly
    1.05
     aplicable
    0.99
     setti
    0.98
    それでは
    0.98
     erwäh
    0.96
    哎
    0.96
    Activations Density 0.011%

    No Known Activations