© 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 - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 27929
    Prev
    Next
    INDEX
    Explanations

    code-related elements such as functions, variables, and programming constructs

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

    code snippets with special characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tuple
    -0.73
    basicConfig
    -0.71
    combination
    -0.71
    memcmp
    -0.70
    NSDictionary
    -0.70
     tuple
    -0.70
     combination
    -0.70
    awtextra
    -0.68
     pair
    -0.68
     Combination
    -0.68
    POSITIVE LOGITS
    SourceChecksum
    0.47
     pelota
    0.46
    erstein
    0.44
    expandindo
    0.44
    !("{}",
    0.42
     BioLib
    0.41
     sola
    0.41
    UseVisualStyle
    0.39
    WritableDatabase
    0.39
    velu
    0.38
    Activations Density 0.247%

    No Known Activations