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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 28-GEMMASCOPE-RES-16K
    11. 6859
    Prev
    Next
    INDEX
    Explanations

    modal verbs, particularly those indicating future certainty or possibility

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

    Comes before "will" or "'ll"

    np_acts-logits-general · gemini-2.0-flash

    will be, will have, will need

    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
    身
    -0.31
     оригинала
    -0.27
     presente
    -0.24
     mention
    -0.24
     à
    -0.23
    ıştı
    -0.23
    ówczas
    -0.23
    ➋
    -0.23
     postale
    -0.23
     present
    -0.22
    POSITIVE LOGITS
    GEBURTSDATUM
    0.76
     doubtless
    0.74
    localctx
    0.71
     probabilmente
    0.69
     probably
    0.69
     waarschijnlijk
    0.68
    expandindo
    0.68
     undoubtedly
    0.66
    principalTable
    0.65
    queryInterface
    0.65
    Activations Density 0.328%

    No Known Activations