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 - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1191
    Prev
    Next
    INDEX
    Explanations

    scientific terms and elements related to stellar evolution and nucleosynthesis

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

    chemical elements and their abundance in the context of astrophysics

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

    chemical compositions and stellar evolution

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    IntoConstraints
    -0.61
    arXiv
    -0.46
     recons
    -0.46
    EndInit
    -0.45
     Américas
    -0.43
    parable
    -0.41
     mí
    -0.40
    ıştı
    -0.40
     PackageManager
    -0.39
     Cyrus
    -0.39
    POSITIVE LOGITS
     متعلقه
    0.69
    OGND
    0.67
    Hochspringen
    0.67
    AxisAlignment
    0.67
    GEBURTSDATUM
    0.67
    GraphicsUnit
    0.65
    SBATCH
    0.64
    ρίς
    0.60
    Vidite
    0.60
    setVerticalGroup
    0.59
    Activations Density 0.302%

    No Known Activations