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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 11004
    Prev
    Next
    INDEX
    Explanations

    references to blood and its related components or conditions

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

    blood and associated medical conditions

    np_acts-logits-general · gemini-2.0-flash

    blood-related medical terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    writeField
    -0.76
     OMITBAD
    -0.71
     >=",
    -0.69
     lccn
    -0.67
    hals
    -0.64
    enumi
    -0.63
    iffel
    -0.63
    berdayakan
    -0.63
    NOPQRST
    -0.63
     quelcon
    -0.62
    POSITIVE LOGITS
     pressure
    0.80
     sugar
    0.80
    sugar
    0.69
    pressure
    0.62
    thirsty
    0.61
     Sugar
    0.61
     vessel
    0.61
     vessels
    0.60
     Pressure
    0.59
    stained
    0.59
    Activations Density 0.143%

    No Known Activations