© 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. 7474
    Prev
    Next
    INDEX
    Explanations

    adding, addition

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

    The neuron strongly activates on relatively rare, content-dense words (e.g. uncommon nouns, verbs, and adjectives), effectively flagging high-information tokens.

    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
    renderCamera
    0.54
    வாய்
    0.50
     laufen
    0.50
     руковод
    0.49
    ľ
    0.49
    வாய
    0.49
     executar
    0.49
    晒
    0.48
     رہا
    0.48
    पल
    0.47
    POSITIVE LOGITS
     additions
    3.39
     addition
    3.25
     добавля
    3.22
     추가
    3.20
     added
    3.19
    追加
    3.19
    添加
    3.19
    を追加
    3.17
     добав
    3.16
     adding
    3.11
    Activations Density 1.191%

    No Known Activations