© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 65330
    Prev
    Next
    INDEX
    Explanations

    terms of endearment and pity

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

    This neuron responds strongly to affectionate or emotionally laden address terms and exclamations (e.g. darling, beloved, dear, sweet) that convey intimacy or strong feeling.

    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
    冷冷
    -1.41
    
    -1.35
     berbagai
    -1.31
     broader
    -1.27
     are
    -1.26
     però
    -1.26
     deeper
    -1.25
    led
    -1.24
    izational
    -1.23
     trovar
    -1.23
    POSITIVE LOGITS
     alliés
    1.29
    胀
    1.29
    }});
    1.27
    線が
    1.26
    )))));
    1.24
     можете
    1.23
    Only
    1.23
    <h2>
    1.23
    Shortly
    1.21
    While
    1.20
    Activations Density 0.039%

    No Known Activations