© 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. 96657
    Prev
    Next
    INDEX
    Explanations

    user instructions and navigational steps in software

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

    sequence then

    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
     nahilalakip
    -0.51
    __':
    
    -0.48
    IsMutable
    -0.48
    Географиясе
    -0.48
     independence
    -0.47
    المناصب
    -0.46
    HandlerContext
    -0.46
     fileSize
    -0.46
    __((
    -0.46
    WriteBarrier
    -0.45
    POSITIVE LOGITS
    then
    0.46
     then
    0.45
    Then
    0.43
     Then
    0.42
    THEN
    0.37
     snippetHide
    0.37
     voila
    0.36
    scriptcase
    0.35
     luego
    0.35
     simplifié
    0.35
    Activations Density 0.023%

    No Known Activations