© 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. 10-GEMMASCOPE-RES-131K
    7. 101472
    Prev
    Next
    INDEX
    Explanations

    prepositional phrases

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

    This neuron strongly activates on two-word comparative or relational phrases combining an adjective or verb with a preposition (e.g. “similar to,” “different from,” “compared with,” “equivalent to”).

    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
     "
    -2.09
     Pokud
    -1.81
    งิน
    -1.59
     služby
    -1.55
     deliciosos
    -1.54
     فهي
    -1.52
    谙
    -1.52
     ".
    -1.50
     Дру
    -1.48
     informée
    -1.48
    POSITIVE LOGITS
     other
    2.23
     others
    2.13
     semasa
    1.78
     dibanding
    1.66
     Rückseite
    1.63
    0
    1.60
    </h1>
    1.57
    }
    1.53
     behulp
    1.51
     so
    1.48
    Activations Density 0.007%

    No Known Activations