© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 28-GEMMASCOPE-RES-16K
    6. 4136
    Prev
    Next
    INDEX
    Explanations

    contractions and negations in casual speech

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

    Follows certain words ("a", "on", "let")

    np_acts-logits-general · gemini-2.0-flash

    edit annotation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     AssemblyCulture
    -0.59
    RenderAtEndOf
    -0.57
    setupUi
    -0.50
     pescoço
    -0.50
    ValueStyle
    -0.49
    ↹
    -0.48
    aarrggbb
    -0.47
     bezpieczeństwa
    -0.47
     становника
    -0.45
     AspNetCore
    -0.45
    POSITIVE LOGITS
    <blockquote>
    0.64
     Edited
    0.57
     yeah
    0.56
    Edited
    0.55
    Edit
    0.54
    @
    0.54
    edited
    0.52
    EDIT
    0.51
     edited
    0.50
     edit
    0.49
    Activations Density 0.442%

    No Known Activations