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-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 3715
    Prev
    Next
    INDEX
    Explanations

    elements related to layout and structure in programming or graphical user interfaces

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

    Code snippets or code-related structures

    np_acts-logits-general · gemini-2.0-flash

    programming code snippets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
    IntoConstraints
    -0.65
    Personendaten
    -0.57
    expandindo
    -0.52
     disambiguazione
    -0.49
     <=",
    -0.49
     IBOutlet
    -0.45
     ModelExpression
    -0.43
    cadi
    -0.42
    WEBPACK
    -0.41
     يتيمه
    -0.40
    POSITIVE LOGITS
    EndInit
    0.60
    évaluateur
    0.52
    ActionCreators
    0.45
    FieldBuilder
    0.43
    BeginInit
    0.40
    scriptcase
    0.40
    internal
    0.40
     Innen
    0.39
    uxxxx
    0.38
    Дереккөздер
    0.38
    Activations Density 0.285%

    No Known Activations