© 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. 104409
    Prev
    Next
    INDEX
    Explanations

    This file is part of

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

    The neuron strongly fires on the first token of frequent multiword expressions (i.e. the lead word of common two‐word phrases).

    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
    </b>
    -1.58
     vlož
    -1.34
    près
    -1.34
    🍡
    -1.30
    Already
    -1.30
     svoju
    -1.30
     variété
    -1.27
     också
    -1.24
     života
    -1.23
    cinq
    -1.23
    POSITIVE LOGITS
     there
    2.36
     posibles
    1.47
    educt
    1.44
    there
    1.40
     list
    1.39
    urnum
    1.35
     competir
    1.34
    itées
    1.31
    ();*/
    1.30
     ello
    1.30
    Activations Density 0.105%

    No Known Activations