© 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. 3261
    Prev
    Next
    INDEX
    Explanations

    this neuron seems to activate on assorted words and symbols in research papers, especially around citations, equations, and section headings

    oai_token-act-pair · gemini-2.0-flash

    capitalized words starting with a specific letter

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

    academic writing

    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
     Shakspeare
    -1.28
     Majefty
    -1.21
     Monfieur
    -1.20
     Houſe
    -1.16
     Shaksp
    -1.14
     pleaſure
    -1.13
     purpoſe
    -1.10
     Efq
    -1.10
     Reſ
    -1.09
     itſelf
    -1.09
    POSITIVE LOGITS
     Sa
    0.60
     Le
    0.59
     De
    0.57
     La
    0.55
     Ho
    0.54
     Ba
    0.54
    ']['
    0.53
     Mer
    0.53
     ++;
    0.52
     Me
    0.52
    Activations Density 7.121%

    No Known Activations