© 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. 74927
    Prev
    Next
    INDEX
    Explanations

    implementation of

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

    The neuron activates on occurrences of “implement” (e.g. implementation, implementing, implemented) and its close morphological variants.

    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
    </h1>
    -2.48
    '
    -2.48
     votre
    -2.19
     immense
    -2.14
     {
    -2.14
     approving
    -2.13
     pinpoint
    -2.11
     converse
    -2.06
     strangle
    -2.06
    偏偏
    -2.05
    POSITIVE LOGITS
    ஂ
    3.13
    衎
    3.00
    玗
    2.80
    啣
    2.73
    玕
    2.72
    鈁
    2.72
    𓆜
    2.69
     sement
    2.61
     asma
    2.59
    )”
    2.53
    Activations Density 0.021%

    No Known Activations