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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-2B
    3. 19-GEMMASCOPE-TRANSCODER-16K
    4. 7060
    Prev
    Next
    INDEX
    Explanations

    This neuron seems to activate inconsistently for different categories, possibly related to gestures, shapes or ways of creating something.

    oai_token-act-pair · gemini-2.0-flash

    letter

    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
    MLLoader
    -0.83
    AddTagHelper
    -0.81
    migrationBuilder
    -0.80
    IVEREF
    -0.80
     initComponents
    -0.73
    InstanceState
    -0.72
     الحره
    -0.71
    脚注の使い方
    -0.71
    DockStyle
    -0.69
     Chwiliwch
    -0.69
    POSITIVE LOGITS
     letter
    0.97
     letters
    0.92
     Letter
    0.81
    Letter
    0.78
     LETTER
    0.78
     shape
    0.77
     Letters
    0.72
    letter
    0.71
     lettre
    0.70
     lettres
    0.69
    Activations Density 1.112%

    No Known Activations