© 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. 11691
    Prev
    Next
    INDEX
    Explanations

    This neuron appears to activate on a mix of copyright notices, numerical sequences, and German sentence structure related to verbs and articles.

    oai_token-act-pair · gemini-2.0-flash

    Code and data files

    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
    ніципа
    -0.67
     come
    -0.67
    BufferException
    -0.65
     che
    -0.65
    terdam
    -0.61
    
    -0.60
     betweenstory
    -0.60
    COME
    -0.59
    come
    -0.59
     cho
    -0.57
    POSITIVE LOGITS
     da
    0.90
     dal
    0.71
    dal
    0.68
    Da
    0.66
     Da
    0.64
     initComponents
    0.60
     dai
    0.59
    da
    0.57
     dalla
    0.53
     AssemblyCulture
    0.51
    Activations Density 0.496%

    No Known Activations