© 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. 22-GEMMASCOPE-RES-131K
    7. 25396
    Prev
    Next
    INDEX
    Explanations

    eigenvalue and eigenvector problems

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

    The neuron is triggered by occurrences of “eigen”-related terms (eigenstates, eigenvalues, eigenvectors).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    references to eigenvalues and eigenvectors in mathematical and physics contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    references to spectral properties and associated states in linear algebra and physics contexts.

    oai_token-act-pair · gpt-5Triggered 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
     is
    -2.58
     vorigen
    -2.06
     {
    -2.03
     where
    -1.98
    S
    -1.91
     einzigen
    -1.88
    缡
    -1.84
     They
    -1.82
    </i>
    -1.81
    摭
    -1.80
    POSITIVE LOGITS
     keramik
    2.61
     protokol
    2.53
    橚
    2.50
     kosme
    2.44
    них
    2.44
     kommentar
    2.41
     enamorado
    2.41
     menyen
    2.41
     himmel
    2.41
     anún
    2.39
    Activations Density 0.008%

    No Known Activations