© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 65118
    Prev
    Next
    INDEX
    Explanations

    technical terms and sequences

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

    The neuron sharply activates on high‐information “content” tokens (proper nouns, technical terms, acronyms, or strong verbs) while ignoring common function words.

    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
    竟然是
    -1.28
    得太
    -1.04
    ’
    -0.95
    Matériau
    -0.94
     ना
    -0.94
     καθώς
    -0.94
    起来的
    -0.93
    aneamente
    -0.92
     reisen
    -0.92
    érieures
    -0.91
    POSITIVE LOGITS
     arrêter
    1.27
     popolari
    1.15
    に入れ
    1.09
    erías
    1.06
     commandés
    1.03
     تھے
    1.02
     patienter
    1.02
     sicherlich
    1.02
     decidedly
    1.02
    とおり
    1.00
    Activations Density 0.094%

    No Known Activations