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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 6120
    Prev
    Next
    INDEX
    Explanations

    elements related to user interface callbacks and actions in code

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

    Strings/code snippets in other languages

    np_acts-logits-general · gemini-2.0-flash

    localization keys or labels

    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
    webElement
    -0.60
     BoxDecoration
    -0.57
    sonaro
    -0.51
    postdata
    -0.51
     sockaddr
    -0.51
    żdy
    -0.50
    ocarp
    -0.50
     كورة
    -0.49
    FailureListener
    -0.48
    poque
    -0.48
    POSITIVE LOGITS
     lbl
    0.69
    numerusform
    0.67
    lbl
    0.66
    InputLabel
    0.64
     phrase
    0.64
    LBL
    0.64
     phrases
    0.63
     title
    0.62
     wording
    0.62
     resourceCulture
    0.61
    Activations Density 0.256%

    No Known Activations