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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 932
    Prev
    Next
    INDEX
    Explanations

    phrases related to selection and handover processes in systems

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

    Text followed by "and"

    np_acts-logits-general · gemini-2.0-flash

    and words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/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.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     coleg
    -0.36
     kirj
    -0.34
     époux
    -0.33
     kuitenkin
    -0.31
    záll
    -0.31
     echip
    -0.30
     tärke
    -0.29
     plads
    -0.29
     autorité
    -0.28
     moitié
    -0.28
    POSITIVE LOGITS
    tagHelperRunner
    0.96
     للاسماء
    0.86
    ValueStyle
    0.80
    transQ
    0.79
    complexContent
    0.79
    IntoConstraints
    0.79
    parsedMessage
    0.75
    ScopeManager
    0.75
    IsContent
    0.73
     kasarigan
    0.71
    Activations Density 0.789%

    No Known Activations