© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 6590
    Prev
    Next
    INDEX
    Explanations

    references to publications and media outlets

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

    references to media, entertainment, and pop culture topics like fashion, celebrities, and awards shows.

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

    references to media publications and brand mastheads, especially in affiliate-disclosure or attribution boilerplate on magazine-style pages.

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

    discourse markers and transitional phrases in informal writing, particularly those that introduce personal commentary or shift between topics.

    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
     SqlCommand
    -0.52
     plastiques
    -0.50
    IBOutlet
    -0.49
    gameserver
    -0.49
    นะครับ
    -0.48
    nictwa
    -0.43
    Tikang
    -0.43
     IBOutlet
    -0.42
    SpringBoot
    -0.42
     gratuites
    -0.41
    POSITIVE LOGITS
    ViewImports
    0.54
     BFF
    0.46
     curated
    0.45
     확인함
    0.44
     Efq
    0.42
    étit
    0.42
     millennial
    0.42
    richTextPanel
    0.41
     boho
    0.39
     Instagram
    0.39
    Activations Density 0.070%

    No Known Activations