© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 15955
    Prev
    Next
    INDEX
    Explanations

    objective scientific entities

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

    The neuron strongly activates on specialized technical and methodological nouns (e.g. “object,” “innovation,” “control,” “groups,” “vocabulary,” “activation,” “graph,” etc.), i.e. it flags scientific/experimental terminology.

    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
    haust
    -0.91
     ferramenta
    -0.86
    短い
    -0.85
    to
    -0.85
    ówczas
    -0.83
    槤
    -0.83
    d
    -0.81
    ＇
    -0.81
    imach
    -0.81
     jetta
    -0.79
    POSITIVE LOGITS
     yani
    1.46
     (
    1.34
     یعنی
    1.08
     czyli
    1.02
    ıt
    0.99
    つまり
    0.98
     Stär
    0.98
     Krá
    0.97
     (=
    0.96
    они
    0.93
    Activations Density 0.121%

    No Known Activations