© 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. 22-GEMMASCOPE-RES-131K
    7. 98292
    Prev
    Next
    INDEX
    Explanations

    quantifiable details and specifications

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

    The neuron strongly activates on quantitative specifications—numbers, units, and measurement terms in product listings.

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

    product descriptions in online marketplace listings, particularly those containing specifications, measurements, prices, and condition details.

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

    descriptive and transactional details typical of e-commerce listings, including specifications, measurements, condition grading, warranties, pricing, and availability statuses.

    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
     they
    -1.51
     their
    -1.35
     then
    -1.16
     launch
    -1.14
     silnika
    -1.09
     started
    -1.09
    satunya
    -1.08
     nadzie
    -1.06
     변경
    -1.05
     already
    -1.05
    POSITIVE LOGITS
    擄
    1.25
    Ľ
    1.24
    rophi
    1.23
    𝑈
    1.19
     ἀ
    1.18
    datable
    1.16
    urator
    1.16
    nessed
    1.16
    💵
    1.15
    睑
    1.15
    Activations Density 0.024%

    No Known Activations