© 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. 22-GEMMASCOPE-RES-131K
    7. 65126
    Prev
    Next
    INDEX
    Explanations

    scientific descriptions and comparisons

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

    The neuron flags tokens that appear in article titles (i.e. the title lines of the scientific papers).

    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
     such
    -1.36
     instead
    -1.25
     then
    -1.23
     bahía
    -1.14
    ainya
    -1.13
    就可以了
    -1.13
    Wanneer
    -1.09
     simply
    -1.09
     terraza
    -1.06
     chimenea
    -1.05
    POSITIVE LOGITS
     unterschied
    1.27
     Different
    1.27
     Both
    1.19
    xtures
    1.18
    Different
    1.14
    ksesta
    1.11
     Some
    1.11
    不同
    1.06
     Neither
    1.06
     betrug
    1.04
    Activations Density 0.387%

    No Known Activations