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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 6108
    Prev
    Next
    INDEX
    Explanations

    words associated with specific individuals or entities

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

    proper nouns and place names, particularly those with specific phonetic patterns or syllable structures.

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

    non-English proper nouns and transliterated names, particularly from Indian, African, and other non-Western languages.

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

    tokens with reduplicated or repeated short syllables/letter sequences, often appearing in proper nouns, abbreviations, or technical strings.

    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
    webElementXpaths
    -0.64
     يتيمه
    -0.61
     autorytatywna
    -0.59
    UrlResolution
    -0.58
     GenerationType
    -0.57
    دانشنامهٔ
    -0.57
     CreateTagHelper
    -0.55
    UnsafeEnabled
    -0.54
    StoryboardSegue
    -0.52
    GEBURTSDATUM
    -0.52
    POSITIVE LOGITS
    nestjs
    0.38
     TUT
    0.36
     bubuk
    0.36
    jojo
    0.35
     Gag
    0.34
    Jojo
    0.33
     antemano
    0.33
     gag
    0.33
     faldas
    0.32
    Bab
    0.32
    Activations Density 0.358%

    No Known Activations