© 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-GEMMASCOPE-TRANSCODER-16K
    7. 12673
    Prev
    Next
    INDEX
    Explanations

    descriptions of cars or car-related places, people, or incidents

    oai_token-act-pair · gemini-2.0-flash

    say car types

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Driving
    -0.75
    Driving
    -0.72
     driving
    -0.71
     drive
    -0.63
     drives
    -0.62
     drove
    -0.61
    driving
    -0.60
     ModelExpression
    -0.56
    Drive
    -0.55
     Drives
    -0.54
    POSITIVE LOGITS
    
    0.72
     pickup
    0.68
     ProtoMessage
    0.67
     sport
    0.66
    Composable
    0.66
    ftagPool
    0.66
     Charger
    0.65
     PICK
    0.64
     JpaRepository
    0.63
     pick
    0.62
    Activations Density 3.739%

    No Known Activations