© 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
    4. Residual Stream - 16k
    5. 33-GEMMASCOPE-RES-16K
    6. 764
    Prev
    Next
    INDEX
    Explanations

    structured data formats and programming syntax, particularly related to regions and entries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.33.hook_resid_post
    Hook Layer
    33
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     StatefulWidget
    -0.60
    Personensuche
    -0.49
     okuyayım
    -0.46
     vostri
    -0.42
     lettore
    -0.42
     StatelessWidget
    -0.41
    muñeca
    -0.40
    matchCondition
    -0.40
    hObject
    -0.40
     onBind
    -0.40
    POSITIVE LOGITS
     right
    0.43
     antigüedad
    0.42
    Bree
    0.41
     Right
    0.41
    OrBuilder
    0.41
    :][
    0.41
     Bioaccumulative
    0.40
    новништво
    0.40
    تقاوى
    0.40
    EndContext
    0.40
    Activations Density 0.811%

    No Known Activations