© 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. 40021
    Prev
    Next
    INDEX
    Explanations

    this followed by a noun

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

    The neuron fires on words appearing right at the start of a sentence or paragraph (i.e. the first token after a break).

    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
    创始人
    -1.01
    espejo
    -0.98
    carteles
    -0.91
    FUENTE
    -0.91
    siempre
    -0.91
    mejores
    -0.91
    を利用
    -0.89
    以前的
    -0.88
    niños
    -0.86
     właścic
    -0.86
    POSITIVE LOGITS
     overall
    1.46
     Overall
    1.34
    Overall
    1.32
    overall
    1.09
     all
    0.97
     this
    0.94
    OVERALL
    0.92
    大会
    0.91
     Gesam
    0.86
    ↵↵
    0.85
    Activations Density 0.068%

    No Known Activations