Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 38-GEMMASCOPE-RES-16K
    6. 7448
    Prev
    Next
    INDEX
    Explanations

    legal terminology and references to courtroom procedures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    featureID
    -0.42
    COMPONENT
    -0.38
    composite
    -0.38
    stateMutability
    -0.38
     composite
    -0.38
    KURZBESCHREIBUNG
    -0.37
    Composite
    -0.36
     Composite
    -0.36
    ysacchar
    -0.35
    canth
    -0.35
    POSITIVE LOGITS
     council
    3.38
     Council
    3.34
    Council
    3.16
     COUNCIL
    3.02
    council
    2.94
     councils
    2.63
     Councils
    2.50
     Coun
    2.22
    UNCIL
    2.22
    ouncil
    2.17
    Activations Density 0.575%

    No Known Activations