© 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. 73674
    Prev
    Next
    INDEX
    Explanations

    storage and placement

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

    The neuron responds to spatial‐relation words—prepositions and directional markers (in, on, into, out, by, etc.) that specify where or how something is placed.

    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
    了一声
    -1.07
    ricus
    -1.03
    ULER
    -0.99
     LABORATORY
    -0.86
    賀状
    -0.84
    컸
    -0.84
    ěli
    -0.84
     hosszú
    -0.83
     الصحة
    -0.82
     instalado
    -0.82
    POSITIVE LOGITS
     out
    4.22
     outta
    2.02
    out
    1.90
     Out
    1.90
    Out
    1.84
    OutOf
    1.72
     OUT
    1.63
     away
    1.61
     unseen
    1.45
     fuera
    1.42
    Activations Density 0.035%

    No Known Activations