© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 12957
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on words that denote offering help, support, or services (e.g. assist, aid, program, provide).

    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
     무슨
    0.63
    ಕ್ತಿಯ
    0.57
    らず
    0.56
     ভীষণ
    0.56
     supper
    0.55
    捐
    0.55
     একটু
    0.54
    饭
    0.54
     forgetting
    0.53
     fullback
    0.53
    POSITIVE LOGITS
     membantu
    0.93
     ayudan
    0.91
     facilitate
    0.90
     enable
    0.88
     facilitates
    0.86
     تساعد
    0.86
     umož
    0.85
     umożliw
    0.84
    協助
    0.84
     помогает
    0.84
    Activations Density 0.284%

    No Known Activations