© 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. 10-GEMMASCOPE-RES-131K
    7. 127995
    Prev
    Next
    INDEX
    Explanations

    FYI or Q for questions

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

    The neuron fires on attention-getting or “heads-up” preambles (e.g. “FYI,” “heads up”) at the start of a message.

    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
     Thus
    -1.74
    ↵
    -1.73
     offiziellen
    -1.70
     eneste
    -1.70
    面對
    -1.68
     falschen
    -1.66
     erfolgreichen
    -1.66
    噥
    -1.64
    1
    -1.62
    叽
    -1.60
    POSITIVE LOGITS
     londres
    1.73
     MessageBoxIcon
    1.73
    蓿
    1.69
    曁
    1.68
    仍在
    1.67
    🫃
    1.66
     itong
    1.66
     llenos
    1.66
    certainly
    1.63
    仍然
    1.63
    Activations Density 0.005%

    No Known Activations