© 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. 59970
    Prev
    Next
    INDEX
    Explanations

    Thanks

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

    The neuron fires on the “A:” answer header and on the trailing “Thanks.” signature (i.e. markers that denote the start of an answer or the end‐of‐post sign‐off).

    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
     by
    -1.36
    igraphy
    -1.34
    다는
    -1.20
     CERTAIN
    -1.19
     pensieri
    -1.19
     messaggi
    -1.18
    heiros
    -1.18
     Otras
    -1.16
     of
    -1.16
     densité
    -1.13
    POSITIVE LOGITS
     who
    1.63
     considerar
    1.37
    任何
    1.25
    鉺
    1.23
     mirar
    1.23
    Their
    1.20
     республики
    1.20
    ༑
    1.19
    ectores
    1.18
    schwar
    1.18
    Activations Density 0.309%

    No Known Activations