© 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. 28332
    Prev
    Next
    INDEX
    Explanations

    to understand

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

    The neuron strongly activates on direct reader‐addressing and imperative (call‐to‐action) language (e.g. “you,” “should,” “take,” “read”).

    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
    findOneAndUpdate
    -0.99
    inglés
    -0.98
    pasti
    -0.98
    pruch
    -0.94
     acord
    -0.93
    Verweise
    -0.92
    fordshire
    -0.89
     brin
    -0.89
     saham
    -0.87
    zungs
    -0.87
    POSITIVE LOGITS
     visit
    1.28
     examination
    1.15
     through
    1.13
     analizar
    1.05
     examining
    1.04
     flexibilidad
    1.04
     diferite
    1.00
     inspection
    0.99
     transparencia
    0.98
     abiertos
    0.97
    Activations Density 0.085%

    No Known Activations