© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 14305
    Prev
    Next
    INDEX
    Explanations

    relative and connecting words

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    </sup>
    -1.28
    </em>
    -1.27
    3
    -1.20
    </b>
    -1.18
    9
    -1.17
    .
    -1.14
    Zdrav
    -1.12
    8
    -1.12
    Where
    -1.09
    もちろん
    -1.09
    POSITIVE LOGITS
     that
    1.43
     which
    1.39
     its
    1.38
     who
    1.31
     οποία
    1.26
    itself
    1.19
    cables
    1.18
    dekat
    1.16
     οποίο
    1.16
     whose
    1.14
    Activations Density 0.025%

    No Known Activations