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 - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 855
    Prev
    Next
    INDEX
    Explanations

    The presence of specific numeric or coded information

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

    numeric values and identifiers

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

    phone number or label

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     key
    -0.46
    /
    -0.45
    mybatisplus
    -0.43
     per
    -0.43
    illeur
    -0.42
    ші
    -0.41
     den
    -0.40
     denn
    -0.39
    AZIONI
    -0.38
     amour
    -0.38
    POSITIVE LOGITS
    EDEFAULT
    0.92
     Phone
    0.92
     ExecuteAsync
    0.91
     ostavi
    0.90
     phone
    0.86
     PHONE
    0.85
    Personensuche
    0.84
     Telephone
    0.84
    Phone
    0.83
    DockStyle
    0.81
    Activations Density 0.018%

    No Known Activations