© 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. 24-GEMMASCOPE-TRANSCODER-16K
    7. 14727
    Prev
    Next
    INDEX
    Explanations

    This appears to be a multilingual neuron selecting for legal jargon, german modal verbs and words containing the substring "ri".

    oai_token-act-pair · gemini-2.0-flash

    Foreign words and code

    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
     pinulongan
    -0.55
    ornal
    -0.51
     consultato
    -0.50
    obacter
    -0.49
    gnation
    -0.49
     Sim
    -0.49
    abin
    -0.48
    osoma
    -0.47
     sim
    -0.47
     LoginForm
    -0.47
    POSITIVE LOGITS
     kasarigan
    0.80
    uxxxx
    0.68
     BoxFit
    0.67
     perfons
    0.67
    TypedDataSet
    0.66
     myſelf
    0.65
    IsContent
    0.64
     purpoſe
    0.63
     houſe
    0.60
     CreateTagHelper
    0.59
    Activations Density 0.815%

    No Known Activations