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

    Neuronpedia

    APICircuit 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. 12682
    Prev
    Next
    INDEX
    Explanations

    references to entities, their characteristics, and related numerical data or results

    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
    ,
    -0.52
    ↵
    -0.35
     (
    -0.30
     "
    -0.30
    <eos>
    -0.29
    2
    -0.27
    1
    -0.25
    left
    -0.25
    )
    -0.25
     running
    -0.25
    POSITIVE LOGITS
    ſſung
    1.02
    rungsseite
    1.02
    niſſe
    1.01
    <unused43>
    1.01
    <unused79>
    1.01
    <unused28>
    1.00
    <unused41>
    1.00
    <unused14>
    1.00
    <pad>
    1.00
    <unused8>
    1.00
    Activations Density 1.324%

    No Known Activations