Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 16311
    Prev
    Next
    INDEX
    Explanations

    references to specific technology or electronic devices

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

    A period that marks the end of a factual statement, often preceding a transition to a new topic or section of text.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    了口气
    -0.34
     complémentaires
    -0.33
    مصادر
    -0.33
     ireo
    -0.33
    temu
    -0.32
     lest
    -0.31
     Direktur
    -0.31
    private
    -0.30
    getRuntime
    -0.29
    RoleName
    -0.29
    POSITIVE LOGITS
    rungsseite
    0.90
    tvguidetime
    0.80
     ujednoznacz
    0.75
    帖最后由
    0.73
     ddelweddau
    0.67
    Personendaten
    0.66
     ModelExpression
    0.65
     surla
    0.65
     ब्रेकडाउन
    0.60
    GEBURTS
    0.56
    Activations Density 0.064%

    No Known Activations