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. 90042
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and their relationships

    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
    Derbyniad
    -0.50
    MessageTagHelper
    -0.48
    UrlHelper
    -0.47
    oa̍t
    -0.47
    tonode
    -0.46
    anasia
    -0.45
    WithIOException
    -0.45
    دانشنامهٔ
    -0.45
    confirmButton
    -0.44
    awi
    -0.43
    POSITIVE LOGITS
     totalSupply
    0.42
     totales
    0.40
     möj
    0.39
    totalSupply
    0.39
     ouder
    0.35
     MonoBehaviour
    0.35
     relógio
    0.34
    SequentialGroup
    0.34
     inteiro
    0.34
     total
    0.34
    Activations Density 0.134%

    No Known Activations