© 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. 90671
    Prev
    Next
    INDEX
    Explanations

    references to brands or specific products

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

    www addresses and emails

    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
     apsau
    -0.32
     either
    -0.31
     indruk
    -0.31
     mesmo
    -0.31
     verp
    -0.31
     assoluto
    -0.30
     aanv
    -0.29
     başladı
    -0.29
     voorzien
    -0.29
     vernie
    -0.28
    POSITIVE LOGITS
     noDo
    0.69
     Biôgrafia
    0.69
     betweenstory
    0.66
     queſta
    0.65
    parsedMessage
    0.64
     '\\;'
    0.62
    IVEREF
    0.62
     мәкал
    0.59
    səhifə
    0.57
    GEBURTSDATUM
    0.57
    Activations Density 0.028%

    No Known Activations