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-2-2B
    4. Residual Stream - 16k
    5. 5-GEMMASCOPE-RES-16K
    6. 4703
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "knock" and its variations in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    getInteger
    -0.57
    ์ตูน
    -0.56
    ↵↵
    -0.55
     Kurz
    -0.55
    ronesia
    -0.55
    oarece
    -0.55
     C
    -0.54
     cetera
    -0.53
     Burnett
    -0.53
     Seas
    -0.51
    POSITIVE LOGITS
     knock
    2.69
     knocks
    2.50
     knocked
    2.50
     knocking
    2.46
    knock
    2.43
     Knock
    2.37
    Knock
    2.14
    kno
    1.87
     knockdown
    1.77
     Kno
    1.67
    Activations Density 0.079%

    No Known Activations