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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 21-GEMMASCOPE-RES-16K
    6. 15579
    Prev
    Next
    INDEX
    Explanations

    concepts or phrases that describe the emergence or development of ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     estadísticas
    -0.47
     parachoque
    -0.43
     sobě
    -0.41
    AfterEach
    -0.41
     exámenes
    -0.41
     certificación
    -0.40
    muñeca
    -0.40
    orologio
    -0.40
     sanciones
    -0.40
     régulier
    -0.40
    POSITIVE LOGITS
     Idea
    1.37
    Idea
    1.34
     idea
    1.32
     Ideas
    1.30
    idea
    1.30
     IDEA
    1.27
    Ideas
    1.24
     ideas
    1.23
    ideas
    1.20
    IDEA
    1.13
    Activations Density 0.077%

    No Known Activations