© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 110563
    Prev
    Next
    INDEX
    Explanations

    differential or special characters

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

    LaTeX/mathematical notation symbols, particularly dollar signs used to delimit mathematical expressions.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron is primarily triggered by inline emphasis or bold‐formatting markers (e.g. “**…**” or `<strong>…</strong>`).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     शुर
    -1.66
     After
    -1.64
    靄
    -1.63
     supremely
    -1.59
     restricting
    -1.55
     extraordinarily
    -1.53
     blatantly
    -1.49
     Originally
    -1.48
     determining
    -1.48
     cubana
    -1.47
    POSITIVE LOGITS
    🫃
    2.13
    ︲
    1.92
     molti
    1.77
    various
    1.73
    {~
    1.68
    なども
    1.66
    🫅
    1.60
    鰨
    1.59
    several
    1.58
    ྄
    1.58
    Activations Density 0.006%

    No Known Activations