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. 5244
    Prev
    Next
    INDEX
    Explanations

    technical terms and proper nouns related to chemical compounds and systems

    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
    RenderAtEndOf
    -0.96
    InjectAttribute
    -0.90
     BoxFit
    -0.90
    CloseOperation
    -0.88
    IUrlHelper
    -0.86
    WriteBarrier
    -0.86
     jScrollPane
    -0.86
     MainAxisSize
    -0.84
    MigrationBuilder
    -0.82
    rungsseite
    -0.82
    POSITIVE LOGITS
    ige
    0.45
    Whereas
    0.43
    ania
    0.42
    раз
    0.42
    rewind
    0.42
    ity
    0.42
    @{
    0.42
    =-
    0.41
    jah
    0.40
     diagon
    0.40
    Activations Density 0.791%

    No Known Activations