© 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. 19-GEMMASCOPE-TRANSCODER-16K
    7. 9632
    Prev
    Next
    INDEX
    Explanations

    This neuron seems to activate on various code snippets and also words related to Facebook, and other miscellaneous topics, without a clear pattern of what it's looking for

    oai_token-act-pair · gemini-2.0-flash

    code and programming context

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_19/width_16k/average_l0_12
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.19.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
     itſelf
    -1.50
     myſelf
    -1.45
     pleaſure
    -1.37
     themſelves
    -1.36
     Jefus
    -1.30
     Eſ
    -1.30
     houſe
    -1.30
    ſelf
    -1.29
     poffe
    -1.28
     himſelf
    -1.27
    POSITIVE LOGITS
     K
    0.50
     k
    0.40
     #
    0.37
    K
    0.37
     P
    0.36
    -
    0.35
     Y
    0.34
     O
    0.33
     
    0.33
     y
    0.33
    Activations Density 14.297%

    No Known Activations