© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 85195
    Prev
    Next
    INDEX
    Explanations

    proposal / research / request

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

    The neuron specializes in spotting formal proposal‐style discourse—terms around submitting, launching, continuing, or setting deadlines for grants/projects (especially when quantified).

    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
     vernac
    -1.58
     forment
    -1.42
     we
    -1.41
     prodi
    -1.40
     strass
    -1.36
     psychi
    -1.30
     galo
    -1.30
     nage
    -1.29
     veng
    -1.29
    If
    -1.29
    POSITIVE LOGITS
     musíte
    1.45
     or
    1.42
     into
    1.41
     about
    1.31
    ruguay
    1.30
     капю
    1.26
    vět
    1.26
     embarazada
    1.25
    仅仅
    1.23
     schrijven
    1.23
    Activations Density 0.246%

    No Known Activations