© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 12-RES-MATRYOSHKA-DC
    7. 20313
    Prev
    Next
    INDEX
    Explanations

    instances of the word "sometimes."

    oai_token-act-pair · gpt-4o-mini
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.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.12.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    chanind/pile-uncopyrighted-gemma-1024-abbrv-1B
    Embeds
    IFrame
    Link
    Negative Logits
     source
    -0.66
    源
    -0.66
     co
    -0.60
     Akin
    -0.58
     Source
    -0.58
     fuente
    -0.57
     Fan
    -0.56
     long
    -0.55
    ิลปะ
    -0.55
     SOURCE
    -0.55
    POSITIVE LOGITS
     Majefty
    0.89
    terday
    0.88
     greateſt
    0.87
     nahilalakip
    0.86
    GEBURTSDATUM
    0.85
     ब्रेकडाउन
    0.84
     itſelf
    0.84
     everybody
    0.83
    ChildScrollView
    0.83
     Sext
    0.82
    Activations Density 0.011%

    No Known Activations