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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 114630
    Prev
    Next
    INDEX
    Explanations

    references to deities and divine feminine figures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
     المعيارى
    -0.65
    ToProps
    -0.58
    flashdata
    -0.58
     chapéu
    -0.58
    Datuak
    -0.58
    ⓧ
    -0.57
    ThroughAttribute
    -0.57
    PerformLayout
    -0.54
    TemporalType
    -0.54
     disambiguazione
    -0.53
    POSITIVE LOGITS
     mongoose
    1.04
     Goose
    0.76
    ongoose
    0.76
    goose
    0.73
     goose
    0.73
    Goose
    0.66
    oose
    0.64
    ose
    0.63
    esse
    0.59
    esta
    0.56
    Activations Density 0.002%

    No Known Activations