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. 20-GEMMASCOPE-RES-16K
    6. 7697
    Prev
    Next
    INDEX
    Explanations

    descriptive characteristics of people, particularly their physical features

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

    "hair" or "hair style"

    np_acts-logits-general · gemini-2.0-flash

    hair

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    saraba
    -0.55
    脚注の使い方
    -0.54
    enumi
    -0.52
    mators
    -0.52
    Rujukan
    -0.48
    endregion
    -0.48
    yszer
    -0.47
    namics
    -0.47
    cotch
    -0.47
    openzeppelin
    -0.47
    POSITIVE LOGITS
     hair
    2.19
     Hair
    1.88
    hair
    1.80
     HAIR
    1.79
    Hair
    1.78
     hairs
    1.63
    HAIR
    1.59
     hairstyle
    1.43
     haircut
    1.36
     hairstyles
    1.33
    Activations Density 0.131%

    No Known Activations