© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 104108
    Prev
    Next
    INDEX
    Explanations

    thoughts following certain words

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

    The neuron primarily activates on occurrences of the word “thought” (and its variants like “thoughts,” “thought-action,” “thought stream,” etc.).

    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
    s
    -2.25
    </strong>
    -2.02
    ?’
    -1.98
    ABETH
    -1.98
    when
    -1.94
    with
    -1.90
    A
    -1.87
    which
    -1.80
    by
    -1.77
    When
    -1.76
    POSITIVE LOGITS
    犼
    2.38
    𐄁
    2.08
     italianos
    2.06
    菒
    1.96
    mathrm
    1.94
     suitably
    1.90
    1.89
     erzä
    1.88
     mexicanos
    1.86
    乄
    1.86
    Activations Density 0.015%
    No Known Activations