© 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-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 16299
    Prev
    Next
    INDEX
    Explanations

    specific programming or coding patterns and commands

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    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
    锈钢
    -0.43
    LEADING
    -0.41
    ToDelete
    -0.40
    Liar
    -0.38
     tokopedia
    -0.38
     laine
    -0.37
    зывы
    -0.37
     lawyers
    -0.36
    talking
    -0.35
    AppState
    -0.35
    POSITIVE LOGITS
    addGap
    4.63
    addPreferredGap
    4.13
    addContainerGap
    3.42
    awtextra
    1.27
    GraphicsUnit
    1.05
    IntoConstraints
    1.02
     CreateTagHelper
    0.86
    BufferException
    0.82
    VIAF
    0.80
    GOTREF
    0.77
    Activations Density 0.004%

    No Known Activations