© 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. 34-GEMMASCOPE-RES-131K
    4. 50292
    Prev
    Next
    INDEX
    Explanations

    look out/up/down/back/around

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

    The neuron detects directional adverbial particles (e.g. “up,” “out,” “into,” “down”) when they occur with verbs of looking or motion.

    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
     μυ
    -1.02
    φή
    -0.98
     pouvoirs
    -0.94
    }}_{\
    -0.94
    ۱۶
    -0.93
     collectivités
    -0.93
     abiti
    -0.93
     achats
    -0.91
     communautés
    -0.91
    ální
    -0.89
    POSITIVE LOGITS
     at
    2.59
     see
    1.38
     👀
    1.17
     saw
    1.14
     thấy
    1.08
     sees
    1.07
     into
    1.05
     to
    1.05
    see
    1.03
    👀
    1.00
    Activations Density 0.015%

    No Known Activations