© 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. 11166
    Prev
    Next
    INDEX
    Explanations

    salt

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

    The neuron responds strongly to words containing the substring “sal” (e.g. Salt, Salako, Salcey, salvation).

    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
    hre
    0.59
    檔
    0.56
    ንያ
    0.54
     së
    0.54
    ulin
    0.53
     stir
    0.52
    rev
    0.52
    Fra
    0.51
    θαν
    0.51
    耦
    0.50
    POSITIVE LOGITS
     Sal
    1.07
    Sal
    1.06
     sal
    1.04
     SAL
    1.01
     salinity
    1.01
     Salz
    0.90
    Salta
    0.89
    🧂
    0.86
    샐
    0.82
    sal
    0.82
    Activations Density 0.168%

    No Known Activations