Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 27-GEMMASCOPE-RES-16K
    6. 9664
    Prev
    Next
    INDEX
    Explanations

    code structure elements, particularly related to functions, loops, and conditional statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.27.hook_resid_post
    Hook Layer
    27
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ReusableCell
    -0.65
    Datuak
    -0.56
    orosi
    -0.54
     ويكيميديا
    -0.54
     للمعارف
    -0.49
    istoitu
    -0.49
    DockStyle
    -0.47
     drawstring
    -0.47
     otomatig
    -0.47
    urlencoded
    -0.47
    POSITIVE LOGITS
     if
    1.23
     если
    0.93
     If
    0.92
     якщо
    0.90
    If
    0.88
     jika
    0.86
     eğer
    0.86
     nếu
    0.86
     Eğer
    0.85
    if
    0.84
    Activations Density 0.176%

    No Known Activations