© 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. 12558
    Prev
    Next
    INDEX
    Explanations

    character interactions involving pets, particularly dogs

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

    dogs, horses, or pets

    np_acts-logits-general · gemini-2.0-flash

    dogs and horses

    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
    verwijspagina
    -0.60
    setViewportView
    -0.50
     subsidi
    -0.45
    Demografie
    -0.44
    otels
    -0.44
     CURLOPT
    -0.43
    навли
    -0.42
    textcolor
    -0.42
    旗
    -0.41
     BoxFit
    -0.40
    POSITIVE LOGITS
     adorable
    0.50
    animaux
    0.47
     dog
    0.46
     cutest
    0.45
    parsedMessage
    0.45
    dog
    0.45
    Dog
    0.44
    adorable
    0.43
    🐶
    0.43
     animals
    0.43
    Activations Density 0.133%

    No Known Activations