© 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. 11147
    Prev
    Next
    INDEX
    Explanations

    i don't know why

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

    The neuron responds to interrogative question words (e.g. “how,” “why,” “what”) signaling the start of a question.

    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
     créé
    -1.56
     autres
    -1.48
    𝑾
    -1.38
     celles
    -1.35
    nemy
    -1.34
     unending
    -1.32
     nimmer
    -1.32
    癟
    -1.30
    之为
    -1.30
     puramente
    -1.30
    POSITIVE LOGITS
     they
    1.81
     at
    1.62
     because
    1.46
     when
    1.45
    0
    1.34
     with
    1.28
    3
    1.28
    4
    1.27
    æd
    1.27
     cuán
    1.23
    Activations Density 0.008%

    No Known Activations