© 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. 6351
    Prev
    Next
    INDEX
    Explanations

    terms associated with intimacy and relationships, particularly in a commercial context

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

    after censored words

    np_acts-logits-general · gemini-2.0-flash

    tragedy, procedure, herbs

    np_acts-logits-general · gemini-2.5-flash-lite

    references to explicit or sensitive content—like sexual terms, profanity, or slurs—especially when they’re obfuscated with symbols, asterisks, or heavy punctuation.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
    脚注の使い方
    -0.60
    Chham
    -0.54
    featureID
    -0.52
    Välislingid
    -0.51
    postsleuth
    -0.50
    styleType
    -0.46
    parsedMessage
    -0.46
     Nucl
    -0.45
     CommonModule
    -0.45
    rinth
    -0.44
    POSITIVE LOGITS
     Infórmanos
    0.44
     ब्रेकडाउन
    0.44
     ujednoznacz
    0.44
     跳转至
    0.43
     ModelExpression
    0.40
    certain
    0.39
     Anon
    0.38
     betweenstory
    0.37
     informée
    0.37
     Hint
    0.36
    Activations Density 0.495%

    No Known Activations