© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49673
    Prev
    Next
    INDEX
    Explanations

    impact and sound actions

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

    The neuron activates on words that describe forceful striking or impact actions (e.g., banging, slamming, whack).

    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
    eslint
    -0.84
    ußer
    -0.77
     staring
    -0.75
     movimentos
    -0.73
     国家
    -0.73
     fornece
    -0.73
    penis
    -0.73
    atherine
    -0.72
    uniqlo
    -0.72
    crack
    -0.71
    POSITIVE LOGITS
     against
    1.48
     on
    1.28
     percussion
    1.11
     drum
    1.11
     banged
    1.10
     drums
    1.09
     banging
    1.09
    敲
    1.00
     fists
    0.99
     rhythms
    0.96
    Activations Density 0.016%

    No Known Activations