© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 14299
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to legal and medical language, particularly in court case descriptions or scientific studies.

    oai_token-act-pair · gemini-2.0-flash

    especially

    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
    #+#
    -0.48
     kotlin
    -0.47
    odles
    -0.45
    aneous
    -0.41
    idopsis
    -0.38
    ungsbedingungen
    -0.38
    Gros
    -0.38
    IBar
    -0.38
     heartbeat
    -0.38
    itudinal
    -0.37
    POSITIVE LOGITS
     especially
    3.38
     particularly
    3.34
    Especially
    3.11
    particularly
    3.06
     Especially
    3.05
    especially
    3.05
     Particularly
    2.98
     especialmente
    2.72
     particularmente
    2.63
     особенно
    2.36
    Activations Density 7.981%

    No Known Activations