© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 32215
    Prev
    Next
    INDEX
    Explanations

    potential candidates

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

    The neuron flags formal evaluative or qualifying terms in technical and academic prose—words like “emerged,” “leading,” “potential,” “candidate,” and “successful” that indicate status or qualification.

    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
    ಇ
    -0.79
     Tumors
    -0.70
    валось
    -0.70
     rents
    -0.70
    unut
    -0.69
     tumors
    -0.69
     GPs
    -0.68
    дю
    -0.66
    gten
    -0.66
    LISA
    -0.66
    POSITIVE LOGITS
     candidates
    7.34
     candidate
    7.19
    candidate
    5.66
     Candidates
    5.44
    candidates
    5.34
    Candidate
    5.31
     Candidate
    5.31
    Candidates
    5.28
     candidatos
    4.34
     candidato
    4.28
    Activations Density 0.178%

    No Known Activations