© 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. 10-GEMMASCOPE-RES-131K
    7. 11186
    Prev
    Next
    INDEX
    Explanations

    value

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

    The neuron fires on occurrences of the word “value” (especially in type or member names like `value_type` or `value`).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     deliciously
    -2.33
     the
    -2.31
     wonderfully
    -2.23
     ridiculously
    -2.23
    㐂
    -2.22
     dando
    -2.22
     beautifully
    -2.19
     delightfully
    -2.16
     hugely
    -2.13
    一下子
    -2.13
    POSITIVE LOGITS
    value
    3.06
     сказала
    2.47
    撷
    2.42
    because
    2.38
    誊
    2.25
    焌
    2.25
    五日
    2.23
    силання
    2.17
     мені
    2.14
    buatan
    2.13
    Activations Density 0.007%

    No Known Activations