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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 12422
    Prev
    Next
    INDEX
    Explanations

    structure and formatting elements typically used in academic or technical documents

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

    document-structure and metadata lines in technical/academic text, such as section headers, titles, filenames, bibliographic labels, and dated author/comment lines.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    section headers and dividers in technical and academic documents, particularly markdown-style underlines made with equals signs or dashes.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    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
     devoir
    -0.45
    </b>
    -0.38
    IsContent
    -0.38
    </strong>
    -0.36
    ParallelGroup
    -0.35
    WriteTagHelper
    -0.35
     gabbana
    -0.34
     econó
    -0.34
     dirinya
    -0.33
     deba
    -0.33
    POSITIVE LOGITS
     CreateTagHelper
    0.93
     informée
    0.88
     autorytatywna
    0.72
     relâche
    0.68
     propOrder
    0.61
     referrerpolicy
    0.59
    +:+
    0.57
     Chwiliwch
    0.57
    abestanden
    0.56
    Autoritní
    0.56
    Activations Density 0.887%

    No Known Activations