© 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. 70463
    Prev
    Next
    INDEX
    Explanations

    halting processes

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

    The neuron fires on occurrences of the word “halt” (and its morphological variants like halting, halted) 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
    贗
    -2.56
     เรา
    -2.52
    っぱり
    -2.38
    دي
    -2.31
    ệc
    -2.31
    鮸
    -2.27
    ẫy
    -2.25
    回来的
    -2.22
    嶮
    -2.20
    
    -2.17
    POSITIVE LOGITS
    );
    2.75
    就
    2.72
    }-
    2.64
    to
    2.58
    1
    2.53
    ;
    2.52
    â
    2.50
     scrutin
    2.48
    没
    2.38
    自己
    2.28
    Activations Density 0.004%

    No Known Activations