© 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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 8915
    Prev
    Next
    INDEX
    Explanations

    terms related to tubular and muscular structures or conditions

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

    "tub" or "mus"

    np_acts-logits-general · gemini-2.0-flash

    tub and musculoskeletal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ArgsConstructor
    -0.57
     فريبيس
    -0.51
     disambiguazione
    -0.49
    Aar
    -0.49
     المعيارى
    -0.48
    autoconfigure
    -0.46
     Extra
    -0.45
    anhão
    -0.44
    ticon
    -0.44
    iesis
    -0.43
    POSITIVE LOGITS
    WebElementEntity
    0.77
    InjectAttribute
    0.69
     Mus
    0.66
    SpringBootTest
    0.64
     Schuyler
    0.62
    glomer
    0.60
    RTLI
    0.60
    tagext
    0.60
     pinulongan
    0.59
     groupBox
    0.59
    Activations Density 0.160%

    No Known Activations