© 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. 24236
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and framework-related terms, particularly in the context of programming code

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

    court citations and claims

    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
    OGND
    -0.80
    LookAnd
    -0.77
    contentLoaded
    -0.75
     للمعارف
    -0.74
    httphttps
    -0.72
     ſever
    -0.70
     Normdatei
    -0.69
    GEBURTSDATUM
    -0.69
     faſt
    -0.68
     وتسجيلات
    -0.66
    POSITIVE LOGITS
     intermediates
    0.31
     volunteers
    0.30
     components
    0.30
     Pro
    0.29
     yine
    0.29
     Ko
    0.29
     events
    0.28
    Http
    0.28
     voluntarios
    0.28
     szak
    0.27
    Activations Density 0.001%

    No Known Activations