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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 15094
    Prev
    Next
    INDEX
    Explanations

    detailed information and descriptions about specific studies or topics

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

    following "of," "on," or "describe"

    np_acts-logits-general · gemini-2.0-flash

    how questions are asked

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     frågan
    -0.50
     financières
    -0.47
     cœurs
    -0.47
     larmes
    -0.45
    anterie
    -0.45
    urnum
    -0.44
    Herzliche
    -0.44
    amblea
    -0.44
    еремо
    -0.44
     Normdatei
    -0.44
    POSITIVE LOGITS
     how
    0.91
     cómo
    0.64
     what
    0.59
     مدى
    0.58
     bagaimana
    0.55
    how
    0.50
     nasıl
    0.49
     كيفية
    0.49
     كيف
    0.49
     hvordan
    0.45
    Activations Density 0.538%

    No Known Activations