© 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. 22-GEMMASCOPE-RES-131K
    7. 35530
    Prev
    Next
    INDEX
    Explanations

    determine whether

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

    This neuron fires strongly on the key content words that open and summarize research‐study sentences—i.e. the main verbs or nouns (like “Induction,” “found,” “aims,” “determine,” “investigated,” etc.) that kick off scientific‐abstract statements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     this
    -1.15
     or
    -1.08
     any
    -1.05
     financière
    -1.03
     moléculas
    -1.02
     creating
    -0.98
    kket
    -0.97
    ksjon
    -0.96
     those
    -0.95
     Creating
    -0.94
    POSITIVE LOGITS
     beige
    0.96
    ię
    0.94
     unveiling
    0.91
    \
    0.91
     regionale
    0.90
     статье
    0.89
    ease
    0.88
     Dinas
    0.88
     oogen
    0.87
     × 
    0.85
    Activations Density 0.041%

    No Known Activations