© 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. 34-GEMMASCOPE-RES-131K
    7. 106586
    Prev
    Next
    INDEX
    Explanations

    with [descriptor]

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

    The neuron strongly activates on the preposition “with” (and its equivalents, e.g. Portuguese “com”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     see
    -0.88
    があるので
    -0.86
    see
    -0.85
    だけでも
    -0.85
    last
    -0.84
     während
    -0.82
    っかけ
    -0.81
    ingdon
    -0.79
    other
    -0.79
    がありますが
    -0.78
    POSITIVE LOGITS
     sadness
    1.32
     regret
    1.30
     great
    1.20
     deep
    1.10
     heavy
    1.08
     pleasure
    1.00
     humble
    0.98
     mixed
    0.96
    GREAT
    0.92
    Regret
    0.92
    Activations Density 0.022%

    No Known Activations