© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 7587
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to legal sentencing and consequences

    oai_token-act-pair · gemini-2.0-flash

    actions related to assigning penalties or sentences

    np_token-act-pair-logits · gpt-4o-mini

    administer, deliver

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
    tabilis
    -0.48
     formate
    -0.48
    idente
    -0.48
    arques
    -0.47
    MENTE
    -0.47
    Ancien
    -0.47
    ropical
    -0.46
    genous
    -0.46
     pinch
    -0.45
    Simult
    -0.45
    POSITIVE LOGITS
     administered
    1.00
     delivered
    0.95
     bestowed
    0.89
     handed
    0.89
    Datuak
    0.88
     administer
    0.87
     given
    0.85
     besto
    0.84
     deliver
    0.82
     received
    0.82
    Activations Density 3.226%

    No Known Activations