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-2-2B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 5268
    Prev
    Next
    INDEX
    Explanations

    occurrences of the term "signal" in various contexts, particularly in relation to its components and functions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     /\.(
    -0.70
    <h1>
    -0.56
     ProtoMessage
    -0.51
     nahilalakip
    -0.51
    <h2>
    -0.50
     createSlice
    -0.49
     little
    -0.49
     imageNamed
    -0.49
    writeHead
    -0.49
     NgModule
    -0.48
    POSITIVE LOGITS
     تضيفلها
    0.91
    ChildScrollView
    0.73
     Citron
    0.71
    itudinal
    0.70
    usitis
    0.69
    viembre
    0.69
    خرى
    0.68
     Sphinx
    0.68
     للمعارف
    0.68
    ristan
    0.67
    Activations Density 0.095%

    No Known Activations