© 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. 21862
    Prev
    Next
    INDEX
    Explanations

    software components and structures

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

    The neuron flags code identifiers—typically class or module names—that end in “-er” or “-or” (e.g. Serializer, Adapter, Editor, Composer, Decorator).

    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
     ewigen
    -1.42
     wichtigste
    -1.38
     about
    -1.37
     Zustimmung
    -1.34
     Aufenthal
    -1.30
     betreffenden
    -1.28
     Sünden
    -1.27
     December
    -1.27
     Annahme
    -1.26
     belangrijkste
    -1.26
    POSITIVE LOGITS
    which
    1.59
     geen
    1.55
     dieses
    1.52
    z
    1.49
    ваю
    1.49
     inget
    1.44
     directe
    1.41
    ílica
    1.40
    ably
    1.38
    zzlement
    1.38
    Activations Density 0.028%

    No Known Activations