© 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-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 4581
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences that provide evidence or support for a claim or observation, often indicated by words like "based on", "judging from", "as you can tell from", etc.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @aml7990

    evidence related to claims or conclusions drawn from data and observations

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

    Prepositions

    np_acts-logits-general · gemini-2.0-flash

    based on from by

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
    twimg
    -0.50
    ſelf
    -0.49
     ligiloj
    -0.48
    ſelves
    -0.48
    énario
    -0.46
    findpost
    -0.45
    MigrationBuilder
    -0.45
    IsContent
    -0.45
     Monfieur
    -0.44
    UrlResolution
    -0.43
    POSITIVE LOGITS
     evidenced
    0.36
    尾
    0.36
    isert
    0.33
     répondu
    0.33
     التأ
    0.33
    ライズ
    0.33
     виправивши
    0.33
    Искәрмәләр
    0.33
     вывод
    0.33
    səhifə
    0.32
    Activations Density 0.836%

    No Known Activations