© 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. 9547
    Prev
    Next
    INDEX
    Explanations

    tokens likely to be place names, prepositions "in, on, of, at, near, from", and some nouns related to competitions (town, Lough, Wheel)

    oai_token-act-pair · gemini-2.0-flash

    names or terms related to entities or locations

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

    Place names

    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
    +#+
    -0.63
    StoryboardSegue
    -0.61
    Espèce
    -0.50
     հղումներ
    -0.48
    RegistryLite
    -0.47
     popoli
    -0.47
    się
    -0.47
    eseorang
    -0.46
    posób
    -0.45
    Manbalar
    -0.45
    POSITIVE LOGITS
    Zeneca
    0.57
     ſtre
    0.53
     CGSize
    0.53
    tonode
    0.52
     defense
    0.52
    KommentareTeilen
    0.52
    ->___
    0.52
    </caption>
    0.51
     OnInit
    0.51
    crose
    0.51
    Activations Density 0.936%

    No Known Activations