© 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. 5561
    Prev
    Next
    INDEX
    Explanations

    terms related to legal proceedings and administrative roles

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

    legal and administrative terminology and keywords.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    status or metadata markers indicating provisional, temporary, discontinued, or archived items (e.g., interim/acting roles, preliminary/unaudited, reserved/closed/deprecated, plurality opinion, archived links), often appearing in parenthetical or administrative contexts.

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

    words or phrases indicating temporary status, substitution, or preliminary/unfinalized states.

    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
     sumpay
    -0.39
    initComponents
    -0.39
     sportifs
    -0.37
     few
    -0.36
     sürü
    -0.35
    ɵɵ
    -0.35
     knex
    -0.35
     sociétés
    -0.35
     vass
    -0.34
     Pressure
    -0.34
    POSITIVE LOGITS
    帖最后由
    0.58
    transQ
    0.56
    TRAILING
    0.55
     Савезне
    0.54
    twimg
    0.52
    iomanip
    0.51
    +#+
    0.50
    erski
    0.48
    tonode
    0.48
     temporarily
    0.47
    Activations Density 0.562%

    No Known Activations