© 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. 10170
    Prev
    Next
    INDEX
    Explanations

    mentions of websites or online resources

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

    Preceding words of URLs or websites

    np_acts-logits-general · gemini-2.0-flash

    Spektrum of language

    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
     Alongside
    -0.68
    Alright
    -0.65
     badass
    -0.65
    Alongside
    -0.64
    gnancy
    -0.63
    …!
    -0.63
     nuanced
    -0.62
     alongside
    -0.61
     prioritize
    -0.61
     schonmal
    -0.61
    POSITIVE LOGITS
     muß
    0.73
     quelquefois
    0.66
     skall
    0.65
     mußten
    0.57
    faßt
    0.57
     mußte
    0.57
     läßt
    0.56
     daß
    0.56
     occorre
    0.55
     wußte
    0.55
    Activations Density 2.225%

    No Known Activations