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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14551
    Prev
    Next
    INDEX
    Explanations

    terms related to "somatic" conditions and their implications in medical contexts

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

    starting with "om" or "Om"

    np_acts-logits-general · gemini-2.0-flash

    om words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     skolen
    -0.52
     réguli
    -0.48
     espirituales
    -0.46
    WebControls
    -0.46
     juridiques
    -0.45
     løpet
    -0.45
     femininas
    -0.44
     senare
    -0.44
    Webpack
    -0.43
     sauvages
    -0.42
    POSITIVE LOGITS
     defaultstate
    0.84
     wireType
    0.83
    OGND
    0.82
     AIO
    0.76
    __;
    0.75
     Omn
    0.73
     Krom
    0.73
    numerusform
    0.71
    ConstraintMaker
    0.71
     Yom
    0.71
    Activations Density 0.185%

    No Known Activations