© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 118253
    Prev
    Next
    INDEX
    Explanations

    praising or condemning actions

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

    The neuron detects positive appraisal language—that is, verbs and words expressing praise, commendation, or laudatory sentiment.

    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
    طبيق
    -1.02
     please
    -0.98
    tilage
    -0.95
    儼
    -0.95
    говари
    -0.92
     зимой
    -0.91
    Zubereitung
    -0.91
    breviations
    -0.90
     вопрос
    -0.90
     用户
    -0.89
    POSITIVE LOGITS
     the
    1.60
     efforts
    1.48
     how
    1.28
    worthy
    1.23
    decision
    1.12
     effort
    1.12
     anew
    1.11
     him
    1.09
     şeyi
    1.07
    efforts
    1.07
    Activations Density 0.017%

    No Known Activations