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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 12509
    Prev
    Next
    INDEX
    Explanations

    instances of programming and code structure syntax

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

    code structure and syntax elements like braces, closing parentheses, and control flow keywords in programming languages.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    FunctionFlags
    -0.41
    EndContext
    -0.35
     lanz
    -0.26
     hija
    -0.26
     acus
    -0.25
     Himmel
    -0.25
     desn
    -0.24
    Teléfono
    -0.24
     transparence
    -0.24
    xnn
    -0.24
    POSITIVE LOGITS
     المعيارى
    0.69
    
    0.67
    adpleegd
    0.67
     Infórmanos
    0.66
     müſſen
    0.62
    ulemon
    0.61
    :✨
    0.60
    الحياه
    0.60
    ፩
    0.59
     ſehen
    0.59
    Activations Density 0.127%

    No Known Activations