© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 11824
    Prev
    Next
    INDEX
    Explanations

    prepositional phrases and common verbs

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

    The neuron fires on words that signal procedural or logical relations—e.g. conditionals and action cues like “if,” “for,” “use,” “post,” “moves,” 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_10/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
     adatok
    -1.38
    styp
    -1.37
    ícias
    -1.37
     familien
    -1.33
    Scrollable
    -1.31
    TOWN
    -1.29
     هذا
    -1.29
     คน
    -1.26
     metrów
    -1.25
    鲥
    -1.25
    POSITIVE LOGITS
     what
    2.08
     their
    1.62
     any
    1.55
     aforementioned
    1.45
    ヵ
    1.43
     making
    1.39
    。』
    1.38
     meantime
    1.38
     it
    1.37
     a
    1.36
    Activations Density 0.179%

    No Known Activations