© 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-2-2B
    7. Residual Stream - 16k
    9. 22-GEMMASCOPE-RES-16K
    11. 4800
    Prev
    Next
    INDEX
    Explanations

    modal verbs that indicate possibility, ability, or obligation

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

    Following modal verbs (must, would, could, should)

    np_acts-logits-general · gemini-2.0-flash

    modal verbs followed by actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mourut
    -0.92
     étoit
    -0.91
     auroit
    -0.89
     avoient
    -0.84
     termica
    -0.83
     feroit
    -0.81
     médicaux
    -0.80
     normaux
    -0.80
     étoient
    -0.80
     zijne
    -0.80
    POSITIVE LOGITS
     make
    1.19
     take
    1.18
     have
    1.17
     be
    1.10
     give
    1.10
     create
    1.09
     provide
    1.08
     get
    1.05
     bring
    1.03
     come
    1.03
    Activations Density 1.557%

    No Known Activations