© 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. 34-GEMMASCOPE-RES-131K
    4. 103843
    Prev
    Next
    INDEX
    Explanations

    for + recipient

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

    This neuron detects the preposition “for” when it introduces a target, purpose, or recipient in product or activity descriptions.

    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
    では
    -1.03
    '
    -0.98
    keeper
    -0.96
    それは
    -0.93
    死亡
    -0.93
    เป็น
    -0.92
    альная
    -0.90
    などを
    -0.87
     dahil
    -0.85
    あ
    -0.85
    POSITIVE LOGITS
     adults
    1.57
     everyone
    1.41
     humans
    1.29
     your
    1.27
     every
    1.26
     sale
    1.24
     you
    1.23
     kids
    1.23
     dummies
    1.21
     Erwachsene
    1.16
    Activations Density 0.088%

    No Known Activations