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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 3592
    Prev
    Next
    INDEX
    Explanations

    specific sequences or structures in text data related to programming, particularly involving logic and flow control

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

    Text followed by numbers

    np_acts-logits-general · gemini-2.0-flash

    numbers and list items

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     /\.(
    -0.58
    bootstrapcdn
    -0.52
     enctype
    -0.52
    GenerationType
    -0.51
     NgModule
    -0.49
    WebServlet
    -0.49
    flashdata
    -0.48
     calldata
    -0.48
    fillType
    -0.47
    berton
    -0.47
    POSITIVE LOGITS
     <<<<<<<<<<<<<<
    0.73
     Penga
    0.42
     bø
    0.40
     liksom
    0.40
     membrana
    0.40
     Paglinawan
    0.39
     oprot
    0.38
     مشين
    0.38
    تفصیلات
    0.38
     personnalisée
    0.38
    Activations Density 0.080%

    No Known Activations