© 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. 24-GEMMASCOPE-RES-16K
    6. 1893
    Prev
    Next
    INDEX
    Explanations

    references to the word "can" and expressions of capability or potential

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

    "Can", "must", and other modal verbs

    np_acts-logits-general · gemini-2.0-flash

    auxiliary verbs and modals

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ésultats
    -0.94
    ſicht
    -0.87
    iſchen
    -0.87
    ſehen
    -0.82
    majánló
    -0.82
    ſcher
    -0.82
     ſind
    -0.80
    niſſe
    -0.79
    <unused14>
    -0.79
    <unused79>
    -0.79
    POSITIVE LOGITS
     voici
    0.37
    ,
    0.37
     Getreide
    0.37
    是
    0.36
    は
    0.36
    stdbool
    0.35
    szönöm
    0.34
     Schon
    0.33
    Is
    0.33
     and
    0.32
    Activations Density 0.839%

    No Known Activations