© 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. 24-RES-MATRYOSHKA-DC
    7. 163
    Prev
    Next
    INDEX
    Explanations

    the word "on", but with high activation only when "on" starts a paragraph

    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/blocks.24.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    chanind/pile-uncopyrighted-gemma-1024-abbrv-1B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ModelSerializer
    -0.75
    MENAFN
    -0.66
    HasAnnotation
    -0.65
    enumii
    -0.63
     createState
    -0.59
    Jereo
    -0.58
    ruimte
    -0.56
    jfree
    -0.55
    toMatchSnapshot
    -0.54
     correctes
    -0.54
    POSITIVE LOGITS
     behalf
    1.38
     occasion
    1.02
     top
    0.95
     board
    0.87
     account
    0.79
     occasions
    0.78
     demand
    0.75
     horseback
    0.75
     Monday
    0.74
     Friday
    0.73
    Activations Density 0.980%

    No Known Activations