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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 113489
    Prev
    Next
    INDEX
    Explanations

    lingering or persistent issues

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

    the neuron detects words that convey continued or persistent states (e.g. “linger,” “persist,” “still,” “remain,” 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
     as
    -1.33
            
    -1.20
     again
    -1.13
    andidat
    -1.04
    tection
    -1.04
     gese
    -1.01
     nuovamente
    -1.00
    ustomed
    -0.99
    以后
    -0.98
    http
    -0.98
    POSITIVE LOGITS
     leurs
    1.17
    residual
    1.11
     leur
    1.10
    ведення
    1.09
    在哪里
    1.05
    letzt
    1.05
     tinham
    1.05
    Residual
    1.05
     Tudo
    1.04
     Ağustos
    1.04
    Activations Density 0.099%

    No Known Activations