© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 110507
    Prev
    Next
    INDEX
    Explanations

    hydrochloride and herbicide

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

    scientific or chemical compound names, particularly ending in "ide," "ine," "one," or related to herbicides.

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

    The neuron fires selectively on technical chemical or drug names (e.g. “Triton,” “hydrochloride,” “herbicide,” “metribuzin”), i.e. tokens denoting specific chemical substances.

    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
     the
    -2.73
    我现在
    -2.58
    泺
    -2.52
    翃
    -2.50
    玙
    -2.48
    我不是
    -2.34
     governo
    -2.30
    at
    -2.27
    s
    -2.25
    我没
    -2.22
    POSITIVE LOGITS
    ↵↵
    3.39
    ;
    
    2.83
    '
    2.45
    ⤞
    2.38
    鏵
    2.27
    𖣘
    2.25
    ᝢ
    2.25
     でも
    2.22
    呖
    2.20
    曁
    2.19
    Activations Density 0.021%

    No Known Activations