© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 77231
    Prev
    Next
    INDEX
    Explanations

    forwarding messages

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

    The neuron reliably activates on email thread separator lines and header markers (e.g. dashed “-----” dividers and phrases like “Original Message” or “Forwarded by”), identifying boundaries between forwarded or replied‐to messages.

    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
    veur
    -0.88
    ennung
    -0.82
    说到
    -0.81
    गा
    -0.79
     pronoun
    -0.78
    setVerticalGroup
    -0.76
     Yelp
    -0.76
     downvotes
    -0.75
     ingresso
    -0.74
     nol
    -0.73
    POSITIVE LOGITS
     forwarded
    3.58
     forward
    3.17
     forwarding
    3.16
     forwards
    3.00
     Forward
    2.89
    Forward
    2.72
    forward
    2.69
     FORWARD
    2.58
    FORWARD
    2.41
    fwd
    2.41
    Activations Density 0.063%

    No Known Activations