© 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. 99608
    Prev
    Next
    INDEX
    Explanations

    physical contact with body parts

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

    physical touch interactions between people, particularly hand-holding, grabbing, or grasping.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates on verbs that describe physical contact—especially grasping, holding, hugging, or otherwise touching someone or something.

    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
     targets
    -0.79
    もしろ
    -0.79
    target
    -0.77
     background
    -0.77
    based
    -0.75
    rooms
    -0.73
     target
    -0.73
     holding
    -0.73
     Decken
    -0.73
    äden
    -0.72
    POSITIVE LOGITS
     wrist
    1.38
     shoulders
    1.36
     waist
    1.23
     hand
    1.15
     arm
    1.11
     shoulder
    1.00
     ankles
    0.99
     ankle
    0.98
     wrists
    0.98
     cintura
    0.96
    Activations Density 0.011%

    No Known Activations