© 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. 120010
    Prev
    Next
    INDEX
    Explanations

    here's what

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

    The neuron detects the common “Here’s what…” lead-in phrase (e.g. “Here’s what you need to know,” “Here’s what we say,” “Here’s what you get,” etc.).

    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
     уже
    -1.57
    After
    -1.48
     itself
    -1.46
     Ironically
    -1.33
     after
    -1.31
     as
    -1.27
     During
    -1.23
     After
    -1.23
     zweiten
    -1.23
    In
    -1.22
    POSITIVE LOGITS
     about
    1.41
     lesquels
    1.27
    THEY
    1.26
     THROUGH
    1.22
     borg
    1.20
    ーマル
    1.20
    ্ড
    1.20
    enterio
    1.20
     feveral
    1.19
    鹮
    1.19
    Activations Density 0.037%

    No Known Activations