© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7211
    Prev
    Next
    INDEX
    Explanations

    physical hits

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

    The neuron fires on words describing physical contact actions (especially violent ones like punch, smack, kick, or even benign ones like pat).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uited
    0.59
    有用
    0.58
     tort
    0.52
     surrounding
    0.51
    陷
    0.51
     주변
    0.50
    crim
    0.50
    HttpMethod
    0.50
    Tort
    0.50
     მათი
    0.49
    POSITIVE LOGITS
     slap
    1.05
     slapped
    0.97
     forehead
    0.87
     slapping
    0.82
     smack
    0.82
    脸
    0.79
     face
    0.79
     ضرب
    0.79
     punch
    0.78
    脸上
    0.77
    Activations Density 0.135%

    No Known Activations