© 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. 31-GEMMASCOPE-RES-16K
    11. 6022
    Prev
    Next
    INDEX
    Explanations

    comparative statements regarding quality or superiority

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

    numeric specifications and technical measurements (like screen sizes, resolutions, frame rates, and prices).

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou

    prepositions and function words in technical discussions or product reviews.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    common function words and structural tokens—especially prepositions, articles, pronouns, conjunctions, and punctuation—that shape sentence structure.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Ortsteil
    -0.50
     sogenannte
    -0.49
     Escherichia
    -0.47
     équipements
    -0.47
     Mitteln
    -0.46
     Mitgliedern
    -0.45
     számára
    -0.45
     célèbres
    -0.45
     Mittelpunkt
    -0.45
     leçons
    -0.44
    POSITIVE LOGITS
     noDo
    0.82
     autorytatywna
    0.79
    CloseOperation
    0.77
     fuckin
    0.75
    aarrggbb
    0.71
    whatever
    0.71
     fucking
    0.71
    istoitu
    0.69
    Jeografia
    0.69
     shitty
    0.67
    Activations Density 1.379%

    No Known Activations