© 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-2-2B
    4. Residual Stream - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 2667
    Prev
    Next
    INDEX
    Explanations

    concepts related to physical properties and behaviors of materials and particles in scientific contexts

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

    "can" or "to" followed by a verb

    np_acts-logits-general · gemini-2.0-flash

    ability to display or possess

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 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
    Gemeinsame
    -0.63
    kloped
    -0.61
     transfieras
    -0.60
     NgModule
    -0.60
    TemporalType
    -0.59
     ajudá
    -0.58
    NgModule
    -0.58
    esgue
    -0.57
    UnitTesting
    -0.52
    popper
    -0.52
    POSITIVE LOGITS
     exhibit
    1.16
     exhibits
    1.06
     undergo
    1.01
     undergoes
    0.98
     Exhibit
    0.94
     display
    0.93
     possess
    0.91
     experience
    0.90
     acquire
    0.90
    Exhibit
    0.88
    Activations Density 1.148%

    No Known Activations