© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13823
    Prev
    Next
    INDEX
    Explanations

    separators in text

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

    The neuron detects section or article titles and headings (i.e. standalone title‐like lines).

    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
     truco
    -0.99
    ARANCE
    -0.97
    ÓW
    -0.97
    тувати
    -0.97
     Versammlung
    -0.93
     billede
    -0.92
    FirebaseDatabase
    -0.91
    irnov
    -0.91
    Ӏ
    -0.90
     sportif
    -0.90
    POSITIVE LOGITS
     continuare
    1.21
     also
    1.02
     téměř
    1.02
    賬
    0.96
    也可以
    0.91
    师傅
    0.88
     sostanza
    0.86
     suelos
    0.86
     midday
    0.85
     kostet
    0.84
    Activations Density 0.001%

    No Known Activations