© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 3-RES-MATRYOSHKA-DC
    7. 1771
    Prev
    Next
    INDEX
    Explanations

    The neuron appears to be identifying the word "full" in various contexts, sometimes also capturing words connected with hyphens.

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard
    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
     itſelf
    -1.23
     Efq
    -1.17
     myſelf
    -1.16
     Monfieur
    -1.16
     doubtnut
    -1.12
     Theſe
    -1.09
     Jefus
    -1.09
     pleaſure
    -1.09
     themſelves
    -1.00
     мәкал
    -1.00
    POSITIVE LOGITS
     full
    3.61
     Full
    3.31
    full
    3.25
    Full
    3.23
     FULL
    2.92
    FULL
    2.91
     vollen
    1.98
    フル
    1.98
     pleine
    1.88
     peł
    1.88
    Activations Density 0.111%

    No Known Activations