© 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. 86752
    Prev
    Next
    INDEX
    Explanations

    expecting outcomes

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

    The neuron is triggered by bare‐infinitive verbs—i.e. verbs appearing in “to <verb>” or similar constructions.

    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
     we
    -1.50
     autre
    -1.40
     autres
    -1.32
     can
    -1.31
     outros
    -1.29
     time
    -1.28
    <>
    
    -1.22
    其它
    -1.21
    Ű
    -1.20
     reportedly
    -1.19
    POSITIVE LOGITS
    phòng
    1.35
     odkazy
    1.31
    ายุ
    1.27
    voorbeeld
    1.20
     během
    1.20
     Новости
    1.20
    得好
    1.19
    Whatsapp
    1.19
    lahraga
    1.18
     be
    1.15
    Activations Density 0.099%

    No Known Activations