© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 26388
    Prev
    Next
    INDEX
    Explanations

    Massachusetts insurers, the ex, after some, beatles’

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

    The neuron strongly activates on lowercase hexadecimal digit letters (a–f) in hex‐encoded data.

    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
     of
    -2.52
     or
    -2.19
     (
    -1.52
     for
    -1.49
     I
    -1.44
    D
    -1.43
    R
    -1.37
     this
    -1.31
     over
    -1.31
    providedIn
    -1.30
    POSITIVE LOGITS
     añadido
    1.55
    馓
    1.53
     reduzir
    1.53
    siamo
    1.46
     sfondo
    1.46
     revisa
    1.43
     calitate
    1.41
     băr
    1.41
     această
    1.37
     culoare
    1.37
    Activations Density 0.000%

    No Known Activations