© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2647
    Prev
    Next
    INDEX
    Explanations

    HTML formatting and markup elements

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

    HTML markup tags, especially those used in structural elements like headers and italics/bold formatting.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    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
    twimg
    -0.57
    Portail
    -0.50
     Asbury
    -0.49
     חיצוניים
    -0.49
     autorytatywna
    -0.48
    kuuta
    -0.47
     AssemblyProduct
    -0.47
    ']>;
    -0.47
    FromNib
    -0.46
    믐
    -0.46
    POSITIVE LOGITS
    enderror
    0.52
     nôtre
    0.47
    AccessorTable
    0.42
     OMITBAD
    0.39
    endphp
    0.38
    endpush
    0.36
    FTFY
    0.35
    colgante
    0.35
    aimer
    0.34
    kwij
    0.34
    Activations Density 0.101%

    No Known Activations