© 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. 1625
    Prev
    Next
    INDEX
    Explanations

    specific formatting or structural elements in data or documents

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

    subject or name followed by signature

    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
     FetchType
    -0.75
     OFDb
    -0.68
     queſta
    -0.67
    
    -0.65
     للمعارف
    -0.64
    原始内容存档于
    -0.63
    thansa
    -0.63
     wikipagina
    -0.62
    Tikang
    -0.62
     Administrativna
    -0.60
    POSITIVE LOGITS
    Histor
    0.31
     Hor
    0.30
     uLocal
    0.29
     title
    0.27
     notes
    0.26
     Sú
    0.26
    Slf
    0.26
    WithEmail
    0.26
     Gur
    0.25
     بتاريخ
    0.25
    Activations Density 0.001%

    No Known Activations