© 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. 118109
    Prev
    Next
    INDEX
    Explanations

    Robin Hood

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

    The neuron strongly activates on tokens that are part of named entities or numerical identifiers (e.g. proper names, titles, years, place names).

    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
     Bélgica
    -0.80
     Schn
    -0.78
     shelling
    -0.76
     lavet
    -0.74
     черво
    -0.74
     heti
    -0.73
     Manson
    -0.73
    大人気
    -0.71
    えた
    -0.71
    destroyAll
    -0.70
    POSITIVE LOGITS
     Robin
    1.26
    Robin
    1.13
     Sherwood
    1.10
    robin
    1.10
     Nottingham
    1.09
     outlaw
    1.08
     Maid
    1.03
     robin
    0.93
     Friar
    0.92
    Nottingham
    0.92
    Activations Density 0.012%

    No Known Activations