© 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. 11609
    Prev
    Next
    INDEX
    Explanations

    carbon nanotube

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

    words ending in "ube", particularly in the context of "nanotube" in scientific literature.

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

    This neuron responds to occurrences of the word “nanotube,” picking up its split tokens (e.g. “nanot” and “ube”) in the text.

    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
    '
    -4.00
    i
    -2.78
    い
    -2.63
    And
    -2.53
    ri
    -2.47
    没
    -2.42
    We
    -2.34
    ?”
    -2.33
    si
    -2.27
    อาจ
    -2.27
    POSITIVE LOGITS
    t
    2.64
     şi
    2.47
    and
    2.41
    翃
    2.33
    笵
    2.28
    土産
    2.20
    鯽
    2.19
     oraz
    2.17
     Some
    2.17
    makeConstraints
    2.14
    Activations Density 0.032%

    No Known Activations