© 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. 34-GEMMASCOPE-RES-131K
    7. 107155
    Prev
    Next
    INDEX
    Explanations

    good news

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

    The neuron fires on announcement‐style “news” phrases—especially constructs like “received some … news” or “there’s more … news.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ග
    -0.94
     mondiale
    -0.79
     légales
    -0.76
    rufe
    -0.75
     sentiers
    -0.75
    ೪
    -0.75
     ชุด
    -0.74
     escalier
    -0.73
    ከ
    -0.73
    ItemList
    -0.73
    POSITIVE LOGITS
     good
    5.34
     bad
    4.88
    good
    3.95
    bad
    3.58
     GOOD
    3.11
    Good
    3.08
     buenas
    3.05
    GOOD
    3.00
    Bad
    2.91
     Good
    2.84
    Activations Density 0.075%

    No Known Activations