© 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. 73763
    Prev
    Next
    INDEX
    Explanations

    prepositions and conjunctions

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

    The neuron strongly responds to technical process terms—especially verbs and related nouns describing acts of fixing, removing, repairing, reducing, or isolating (and their associated parameters).

    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
     to
    -0.94
     außer
    -0.90
    íguez
    -0.88
     gearbeitet
    -0.87
     gefüllt
    -0.85
     ponieważ
    -0.85
    Tikang
    -0.85
     домов
    -0.83
     perché
    -0.82
     bedeutend
    -0.81
    POSITIVE LOGITS
     via
    2.03
     through
    1.39
    via
    1.31
     by
    1.20
     berupa
    1.08
     including
    1.07
    Via
    0.97
     pomocí
    0.95
    包括
    0.95
     mediante
    0.94
    Activations Density 0.107%

    No Known Activations