© 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. 34-GEMMASCOPE-RES-131K
    7. 65376
    Prev
    Next
    INDEX
    Explanations

    almost

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

    This neuron specializes in spotting LaTeX‐style mathematical fragments—especially numeric tokens (digits, sub/superscripts) and math delimiters/commands.

    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
     sırasında
    -0.75
     Studio
    -0.72
    trans
    -0.71
    înes
    -0.71
     frukt
    -0.70
     Geister
    -0.69
    getSystemService
    -0.69
    renda
    -0.69
    $-$\\
    -0.67
    xb
    -0.67
    POSITIVE LOGITS
     almost
    1.89
    almost
    1.73
    几乎
    1.66
     почти
    1.66
     prawie
    1.63
    Almost
    1.59
     presque
    1.50
     Almost
    1.49
     casi
    1.47
     hampir
    1.43
    Activations Density 0.031%

    No Known Activations