© 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. 22-RES-MATRYOSHKA-DC
    7. 1604
    Prev
    Next
    INDEX
    Explanations

    uses of the words "for", "in", "on", and "from" that act like prepositions

    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.22.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.22.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
    IVEREF
    -0.56
    fycat
    -0.54
    tsy
    -0.54
    むしろ
    -0.47
    maxAge
    -0.47
    arschijnlijk
    -0.47
    RuleContext
    -0.47
    ządu
    -0.47
    ulemon
    -0.45
    ษัท
    -0.45
    POSITIVE LOGITS
     which
    3.73
    which
    3.05
     WHICH
    2.73
     Which
    2.67
    Which
    2.58
    hich
    2.02
     whom
    2.00
     laquelle
    1.91
     lequel
    1.91
     wich
    1.88
    Activations Density 2.789%

    No Known Activations