© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 111618
    Prev
    Next
    INDEX
    Explanations

    joining or contributing

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

    This neuron detects formal, business‐style announcement or participation verbs (e.g. “submit,” “participating,” “working with,” “teaming”) in professional contexts.

    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
     producenta
    -1.65
     several
    -1.48
     refroid
    -1.46
     verstor
    -1.40
     nastro
    -1.39
     разные
    -1.39
    谂
    -1.39
     amenazas
    -1.38
     mezcla
    -1.38
     same
    -1.38
    POSITIVE LOGITS
     of
    1.91
     and
    1.56
     byz
    1.51
     jose
    1.42
    з
    1.41
     venter
    1.41
    idae
    1.41
    н
    1.27
     at
    1.26
    ribusiness
    1.25
    Activations Density 0.025%

    No Known Activations