© 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. 10-GEMMASCOPE-RES-131K
    7. 103498
    Prev
    Next
    INDEX
    Explanations

    salts and chemical descriptors

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

    references to chemical salts.

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

    The neuron responds to mentions of chemical “salt” compounds, i.e. occurrences of the word “salt(s)” in technical/chemical contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     –
    -2.64
    ’
    -2.52
    7
    -2.47
    .
    -2.23
    ness
    -2.17
    
    -2.08
    an
    -2.06
    <em>
    -2.06
    艄
    -2.05
    ୩
    -2.00
    POSITIVE LOGITS
     TAMBIÉN
    2.31
    嗄
    2.17
     TÉCN
    2.09
    ╿
    2.08
    guigu
    2.06
     TÉCNICA
    1.98
    AISSEE
    1.97
    ：[
    1.96
     prominent
    1.94
     jadwal
    1.94
    Activations Density 0.006%

    No Known Activations