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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 14269
    Prev
    Next
    INDEX
    Explanations

    qualities and characteristics of products, particularly related to durability and effectiveness

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

    nighttime and leaks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    AccessorTable
    -0.45
     accommodation
    -0.36
     beginnetje
    -0.34
     recovered
    -0.34
    awarkan
    -0.32
    RTGC
    -0.31
    createStatement
    -0.31
     Glück
    -0.31
     onOptions
    -0.30
     resourceCulture
    -0.29
    POSITIVE LOGITS
    WebElementEntity
    0.50
     nahilalakip
    0.50
    ỡng
    0.50
     yym
    0.49
     nought
    0.48
    持久
    0.47
    /**
    0.46
    níky
    0.45
    hrs
    0.45
    Хьажоргаш
    0.44
    Activations Density 0.003%

    No Known Activations