© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 3926
    Prev
    Next
    INDEX
    Explanations

    various programming code and related terms

    oai_token-act-pair · gemini-2.0-flash

    code and file types

    np_max-act-logits · gemini-2.0-flash

    uppercase acronyms

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
    virons
    -0.50
    #%%
    -0.47
    hella
    -0.44
    demar
    -0.43
    antino
    -0.43
    RTP
    -0.42
    usha
    -0.42
    reiches
    -0.42
    CPC
    -0.42
    anique
    -0.41
    POSITIVE LOGITS
    s
    0.88
     EconPapers
    0.77
    исленность
    0.69
    DeleteBehavior
    0.69
    HandlerContext
    0.68
    XmlAccessorType
    0.67
     tartalomajánló
    0.66
    ſelves
    0.66
    Personendaten
    0.65
     ſche
    0.63
    Activations Density 6.145%

    No Known Activations