© 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. 72229
    Prev
    Next
    INDEX
    Explanations

    enterprise computing or monitoring tools

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

    The neuron is especially sensitive to technical finance and valuation jargon, such as “true market value,” “mark‐to‐market,” and related terms.

    oai_token-act-pair · o4-miniTriggered 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
     numerosos
    -1.14
    ihnachts
    -1.11
     constituyen
    -1.08
    initialise
    -1.05
     тях
    -1.03
     although
    -1.03
     매우
    -1.02
     여러
    -1.01
     inicios
    -1.00
     compleja
    -0.99
    POSITIVE LOGITS
     solution
    1.23
     like
    1.19
    のように
    1.14
    谩
    1.08
     solutions
    1.01
    のような
    1.00
    ICALS
    1.00
     software
    0.99
    рал
    0.97
    lakukan
    0.96
    Activations Density 0.040%

    No Known Activations