© 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. 39286
    Prev
    Next
    INDEX
    Explanations

    especially for groups or situations

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

    The neuron detects “for X” recommendation phrases—i.e. instances where something is described as being suitable or ideal for a particular group, audience, or occasion.

    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
    ものに
    -1.02
    举动
    -1.02
     alcuni
    -1.00
    も良い
    -0.99
    半天
    -0.97
     incluida
    -0.96
     utilizando
    -0.95
    chiato
    -0.95
    際は
    -0.95
    tedrale
    -0.95
    POSITIVE LOGITS
     especially
    1.67
     because
    1.54
    因為
    1.51
    因为
    1.34
     particularly
    1.33
    任何
    1.27
     any
    1.17
     both
    1.09
     besonders
    1.09
     porque
    1.07
    Activations Density 0.058%

    No Known Activations