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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 76958
    Prev
    Next
    INDEX
    Explanations

    directions of movement

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

    The neuron detects words that indicate spatial or movement directions (e.g. northward, southward, right, northwest).

    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
     forward
    -1.25
     southbound
    -1.23
     northbound
    -1.16
    Forward
    -1.16
     that
    -1.02
    FORWARD
    -1.00
     inward
    -0.99
    نه
    -0.96
    LES
    -0.95
    forward
    -0.95
    POSITIVE LOGITS
     into
    1.36
     away
    1.29
     towards
    1.13
     direction
    1.08
     toward
    1.07
     Direction
    1.01
     نحو
    1.00
     вдоль
    0.98
    toward
    0.93
    away
    0.93
    Activations Density 0.054%

    No Known Activations