© 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. 22-GEMMASCOPE-RES-131K
    4. 67863
    Prev
    Next
    INDEX
    Explanations

    questions in multiple languages

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

    The neuron responds to unusually long runs of whitespace (indentation or large blank gaps).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    
    -1.30
    ᵉ
    -1.28
    testCase
    -1.27
    ︓
    -1.26
    ⃢
    -1.22
    ʰ
    -1.18
     Baños
    -1.17
    avond
    -1.16
     sammen
    -1.14
     conséquent
    -1.14
    POSITIVE LOGITS
     folgt
    1.40
     الحكومة
    1.40
     fünf
    1.34
    都想
    1.30
     jednej
    1.30
    才可以
    1.25
     procedimento
    1.24
     solchen
    1.24
    nové
    1.23
     wonderful
    1.22
    Activations Density 0.010%

    No Known Activations