© 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. 24-GEMMASCOPE-TRANSCODER-16K
    7. 15535
    Prev
    Next
    INDEX
    Explanations

    occurrences of common English prepositions and pronouns, particularly when associated with auxiliary verbs or modal verbs like "allow".

    oai_token-act-pair · gemini-2.0-flash

    us

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.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
     them
    -0.90
    they
    -0.84
     itſelf
    -0.80
     they
    -0.74
    them
    -0.71
    They
    -0.69
    mereka
    -0.69
    让他们
    -0.69
     ſhe
    -0.69
     They
    -0.68
    POSITIVE LOGITS
     us
    2.45
     US
    1.29
     Us
    1.17
     me
    1.14
    us
    0.95
    US
    0.90
    Us
    0.85
     usa
    0.85
     myself
    0.78
     нас
    0.74
    Activations Density 5.822%

    No Known Activations