© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38620
    Prev
    Next
    INDEX
    Explanations

    groups or communities

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

    This neuron fires on the content clauses that follow verbs of liking or joining—i.e. the “what it does” or “that this was a conference of professionals”–type subordinate clauses.

    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
     will
    -1.38
    ৄ
    -1.33
     or
    -1.30
     alberi
    -1.22
    ̀
    -1.20
     menghapus
    -1.15
     would
    -1.10
    uberge
    -1.09
    capacidad
    -1.08
    obut
    -1.08
    POSITIVE LOGITS
     I
    2.06
     my
    1.48
    違います
    1.10
    ツヤ
    1.09
     différentes
    1.09
    さり
    1.05
     wonderful
    1.05
     me
    1.05
     Emple
    1.05
     мне
    1.02
    Activations Density 0.021%

    No Known Activations