Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 11087
    Prev
    Next
    INDEX
    Explanations

    terms related to atomic structure and electron interactions

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

    electric charge and electrons

    np_acts-logits-general · gemini-2.0-flash

    electric charge and charged particles

    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
     Picchu
    -0.47
    ágico
    -0.45
    thmus
    -0.43
    iastes
    -0.43
     ComVisible
    -0.42
    (!__
    -0.40
    CppMethod
    -0.40
    lasia
    -0.39
     descomp
    -0.39
     stand
    -0.39
    POSITIVE LOGITS
     charge
    1.50
     Charge
    1.40
     charges
    1.34
    charge
    1.34
    Charge
    1.33
     charged
    1.29
     Charges
    1.26
     CHARGE
    1.24
    charges
    1.21
    Charges
    1.17
    Activations Density 0.715%

    No Known Activations