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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 24-GEMMASCOPE-TRANSCODER-16K
    7. 3454
    Prev
    Next
    INDEX
    Explanations

    This neuron appears to detect language related to medical fractures or currency amounts with special symbols, and to a lesser extent ordinal numbers as well as voltage and forces.

    oai_token-act-pair · gemini-2.0-flash

    technical and foreign languages

    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
    es
    -0.34
    et
    -0.33
     hai
    -0.32
    ou
    -0.31
     us
    -0.30
     wie
    -0.29
     G
    -0.29
    хова
    -0.29
     Is
    -0.29
     ni
    -0.28
    POSITIVE LOGITS
     pleaſure
    1.34
     myſelf
    1.30
     houſe
    1.27
     itſelf
    1.27
     themſelves
    1.24
     Efq
    1.24
     purpoſe
    1.23
     Theſe
    1.22
    ſelf
    1.21
     Majefty
    1.18
    Activations Density 1.044%

    No Known Activations