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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 11835
    Prev
    Next
    INDEX
    Explanations

    references to Facebook and its associated links or pages

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

    Facebook or FB

    np_acts-logits-general · gemini-2.0-flash

    Facebook and social media

    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
    ThroughAttribute
    -0.55
    providedIn
    -0.50
    omitempty
    -0.47
     Bewußt
    -0.47
    verifyException
    -0.46
     kasarigan
    -0.45
    rrggbb
    -0.45
    __":
    -0.44
    oredCriteria
    -0.43
    выше
    -0.42
    POSITIVE LOGITS
     Facebook
    0.99
    Facebook
    0.90
    facebook
    0.85
     facebook
    0.84
     Instagram
    0.82
     YouTube
    0.82
    Instagram
    0.81
     FACEBOOK
    0.81
    FACEBOOK
    0.76
    YouTube
    0.76
    Activations Density 0.101%

    No Known Activations