Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 8032
    Prev
    Next
    INDEX
    Explanations

    noteworthy achievements and recognitions in the education sector

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ComponentScan
    -0.32
     alignItems
    -0.32
    setMessage
    -0.30
    MessageOf
    -0.30
     JpaRepository
    -0.29
     voisin
    -0.29
     chimiques
    -0.29
    neighbour
    -0.29
     inom
    -0.29
     interesar
    -0.28
    POSITIVE LOGITS
     ब्रेकडाउन
    0.62
     finalist
    0.60
     finalists
    0.59
     selected
    0.57
    bootstrapcdn
    0.55
     selection
    0.53
    cabul
    0.52
    Lugares
    0.52
     Selection
    0.51
     betweenstory
    0.50
    Activations Density 0.199%

    No Known Activations