© 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. 124763
    Prev
    Next
    INDEX
    Explanations

    still remains / keep

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

    The neuron fires on verbs that express continuation or persistence (e.g. remain(s), keep(s)).

    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
    who
    -1.32
     electricidad
    -1.30
    ulipas
    -1.28
     dispositivos
    -1.26
    phère
    -1.25
     لمدة
    -1.24
     });
    
    -1.23
    ‼️
    -1.23
    llllll
    -1.21
    Nonetheless
    -1.21
    POSITIVE LOGITS
    bitkan
    1.48
    ↵↵
    1.47
     namorado
    1.45
     nephew
    1.33
     नए
    1.26
     into
    1.25
     magnitude
    1.24
     иначе
    1.23
     .
    1.23
     لیتر
    1.22
    Activations Density 0.022%

    No Known Activations