© 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. 34-GEMMASCOPE-RES-131K
    7. 47642
    Prev
    Next
    INDEX
    Explanations

    vector or bitmap

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

    The neuron specializes in spotting code‐style markup and technical identifiers—things like HTML/XML tags (e.g. `<script>`, `</script>`, `<link>`) and library/file names (e.g. “Vector”, “Icons”, “react-native-vector-icons”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Tv
    -0.73
    праж
    -0.71
    енный
    -0.70
     Rhodes
    -0.70
    erez
    -0.67
    demics
    -0.67
     Petri
    -0.66
     Pet
    -0.66
    estament
    -0.63
     Singh
    -0.63
    POSITIVE LOGITS
     illustrations
    0.81
    diffuse
    0.78
    deliver
    0.78
     Opin
    0.73
    liefer
    0.73
    TypeError
    0.71
    uste
    0.71
     illustration
    0.70
    FontAwesome
    0.70
    örer
    0.69
    Activations Density 0.062%

    No Known Activations