© 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. 81242
    Prev
    Next
    INDEX
    Explanations

    directional words before a place name

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

    The neuron detects cardinal‐direction modifiers (e.g. “southern,” “southwestern,” “southeast”) used in geographic place names.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    geographic regional descriptors indicating relative position within a larger area using compass directions (e.g., directional adjectives for parts of regions).

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    directional terms indicating geographic locations, particularly "southern", "southeastern", and "eastern".

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
    tilities
    -0.84
     mencionó
    -0.81
    運命
    -0.78
    Referencoj
    -0.77
     fulfilment
    -0.77
     подходя
    -0.76
     Falkland
    -0.76
    Menge
    -0.76
    Síguenos
    -0.76
    Soporte
    -0.75
    POSITIVE LOGITS
    alls
    0.79
     manguera
    0.77
    indahkan
    0.74
    textview
    0.72
     Choosing
    0.71
    istic
    0.71
    óry
    0.70
    spective
    0.69
    搏
    0.69
    Normals
    0.69
    Activations Density 0.037%

    No Known Activations