© 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-2-9B-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 42557
    Prev
    Next
    INDEX
    Explanations

    technical terminology related to molecular modifications and their effects

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

    modification, doping, approval, relaxation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     AssemblyCulture
    -0.73
     disambiguazione
    -0.68
    SharedCtor
    -0.64
    IndentedString
    -0.63
    TagMode
    -0.60
    msgTypes
    -0.59
     للاسماء
    -0.59
     oprot
    -0.59
    InputBorder
    -0.58
    ItemBackground
    -0.58
    POSITIVE LOGITS
    ED
    0.40
    Xna
    0.38
    ded
    0.36
    xed
    0.35
    ed
    0.34
    ‌شده
    0.33
    CppMethod
    0.33
    <s>
    0.33
    <thead>
    0.32
    ized
    0.32
    Activations Density 0.419%

    No Known Activations