© 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. 24952
    Prev
    Next
    INDEX
    Explanations

    phrases indicating spatial relationships or positions

    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
     podes
    -0.79
     orale
    -0.75
     pleaſure
    -0.75
     Pilate
    -0.74
     purpoſe
    -0.71
     transect
    -0.69
     houſe
    -0.67
     Assyrian
    -0.66
     deafness
    -0.66
     ROK
    -0.66
    POSITIVE LOGITS
    cshtml
    0.79
    EARCH
    0.70
     the
    0.69
    Décès
    0.69
     ModelExpression
    0.69
    ientôt
    0.67
    MENTS
    0.66
    aji
    0.65
    roits
    0.65
     them
    0.65
    Activations Density 0.017%

    No Known Activations