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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 22-GEMMASCOPE-RES-16K
    6. 10408
    Prev
    Next
    INDEX
    Explanations

    articles and common verbs indicating actions or roles

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

    Positive adjectives after indefinite articles

    np_acts-logits-general · gemini-2.0-flash

    positive evaluations after articles

    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
    Général
    -0.66
    يكب
    -0.63
     rarest
    -0.58
     darkest
    -0.56
     coldness
    -0.56
    artem
    -0.56
    เร็
    -0.54
    neſs
    -0.52
     blackness
    -0.51
     tropics
    -0.51
    POSITIVE LOGITS
    énario
    0.66
    andReturn
    0.63
    IntoConstraints
    0.62
    rungsseite
    0.58
     promising
    0.57
     fine
    0.56
     excellent
    0.56
     significa
    0.56
    agrama
    0.56
     valid
    0.55
    Activations Density 0.493%

    No Known Activations