© 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. 20405
    Prev
    Next
    INDEX
    Explanations

    for you, myself, yourself

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

    This neuron detects phrases expressing an emotion toward someone else, especially constructions like “[emotion] for [someone]” (e.g. “happy for you,” “excited for us,” “scared for yourself”).

    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
     penyebab
    -1.55
     contribute
    -1.51
    itato
    -1.48
    matian
    -1.46
    apatan
    -1.45
    ༐
    -1.43
    laje
    -1.41
     он
    -1.41
    
    -1.40
     So
    -1.38
    POSITIVE LOGITS
    2
    2.23
    will
    1.69
    ():
    1.65
    4
    1.64
    (!)
    1.60
     tanken
    1.59
     kapas
    1.58
     namun
    1.55
     dekoratif
    1.54
     見る
    1.53
    Activations Density 0.006%

    No Known Activations