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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-2B
    3. 21-GEMMASCOPE-TRANSCODER-16K
    4. 11406
    Prev
    Next
    INDEX
    Explanations

    code-related keywords such as "self" and "user" often found in classes and function definitions, in addition to comments represented by triple quotes

    oai_token-act-pair · gemini-2.0-flash

    words with "self" prefix

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

    archaic words

    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
     
    -0.88
     I
    -0.85
     #
    -0.84
     A
    -0.81
     [
    -0.79
     «
    -0.78
     R
    -0.78
     de
    -0.77
     "
    -0.77
     di
    -0.77
    POSITIVE LOGITS
     myſelf
    1.71
     Majefty
    1.67
     Jefus
    1.65
     Efq
    1.60
     itſelf
    1.58
    ſelf
    1.54
     Monfieur
    1.52
     greateſt
    1.50
     Diſ
    1.50
    ſelves
    1.50
    Activations Density 1.333%

    No Known Activations