© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 15728
    Prev
    Next
    INDEX
    Explanations

    multilingual and mixed content

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

    The neuron responds to non‐alphabetic symbol tokens—especially HTML entities (like “ ”) and multi‐punctuation sequences (e.g. “??”, “.?”, etc.).

    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
    -1.38
     casas
    -1.23
     one
    -1.16
    ;
    -1.16
     
    -1.12
     James
    -1.07
     perpetually
    -1.05
     palmas
    -1.04
     multifaceted
    -1.04
     poignant
    -1.03
    POSITIVE LOGITS
     acestea
    1.40
     AMAZING
    1.22
     insanely
    1.21
     această
    1.19
     aceasta
    1.15
     Begründung
    1.15
     logros
    1.14
     nenhuma
    1.12
     Absicht
    1.11
     SEVER
    1.10
    Activations Density 0.027%

    No Known Activations