© 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. 18-GEMMASCOPE-RES-16K
    11. 12855
    Prev
    Next
    INDEX
    Explanations

    references to object identifiers and function calls within code structures

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

    "obj" followed by a period or parenthesis

    np_acts-logits-general · gemini-2.0-flash

    object properties and indices

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    <strong>
    -1.04
    <em>
    -0.87
    </em>
    -0.72
    5
    -0.70
    <b>
    -0.67
    ingu
    -0.66
    4
    -0.65
    “
    -0.64
    o
    -0.63
    金
    -0.62
    POSITIVE LOGITS
     виправивши
    1.08
    AddTagHelper
    1.03
     himſelf
    1.01
     themſelves
    0.99
     myſelf
    0.99
    httphttps
    0.98
     Theſe
    0.97
    DockStyle
    0.93
     للاسماء
    0.93
    Παραπομπές
    0.91
    Activations Density 0.197%

    No Known Activations