© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 12373
    Prev
    Next
    INDEX
    Explanations

    code syntax patterns and structure in programming language snippets

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

    Color names

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
    ніципалі
    -0.51
     فريبيس
    -0.49
    windowFixed
    -0.48
     disambiguazione
    -0.45
    dafx
    -0.45
     Administrativna
    -0.44
    şört
    -0.44
    ImageField
    -0.44
    RTGC
    -0.44
    তথ্যসূত্র
    -0.43
    POSITIVE LOGITS
     îns
    0.45
    mesini
    0.37
     Werden
    0.35
    XmlAccessType
    0.35
     Stelle
    0.34
     trebui
    0.33
     Nimbus
    0.33
     ķ
    0.32
    IonicModule
    0.32
     følge
    0.31
    Activations Density 0.057%

    No Known Activations