© 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. 12274
    Prev
    Next
    INDEX
    Explanations

    references to images and screenshots related to planning tools

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

    image files and scenes

    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
     disponibilités
    -0.52
     Houſe
    -0.46
     houſe
    -0.46
     ſou
    -0.45
     ſind
    -0.44
     faſt
    -0.44
    jspb
    -0.44
     ſte
    -0.44
    tvguidetime
    -0.43
    ualaikum
    -0.43
    POSITIVE LOGITS
     scenes
    0.42
     Scenes
    0.42
    photo
    0.38
    Datuak
    0.38
     zdjęć
    0.38
     photo
    0.38
     photographs
    0.38
    gettyimages
    0.37
     eind
    0.36
     ilustrasi
    0.36
    Activations Density 0.033%

    No Known Activations