© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 8716
    Prev
    Next
    INDEX
    Explanations

    the presence of numeric values or mathematical expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    raw numerical data, especially at the beginning of lines in technical or code contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    standalone single-character tokens (letters or symbols), often appearing as isolated separators or at the start of a line.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    the digit "1" when it appears at the beginning of a document or line.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    SourceChecksum
    -0.52
    ValueStyle
    -0.43
    httphttps
    -0.40
    MockMvc
    -0.40
     kasarigan
    -0.39
    andExpect
    -0.39
    uxxxx
    -0.39
    AnimationsModule
    -0.39
     disambiguazione
    -0.39
     Савезне
    -0.38
    POSITIVE LOGITS
    #
    0.62
    rrggbb
    0.60
     @"/
    0.57
     mijne
    0.55
    Gön
    0.50
     gebo
    0.50
     zijne
    0.48
     beroep
    0.48
    tagHelperRunner
    0.47
     aspetti
    0.46
    Activations Density 0.206%

    No Known Activations