© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 1578
    Prev
    Next
    INDEX
    Explanations

    references to programming and technology concepts, particularly related to APIs and data handling

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_15/width_16k/average_l0_131
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     duque
    -0.34
     Stirn
    -0.34
     bắt
    -0.33
     Konf
    -0.32
     better
    -0.31
     coloré
    -0.30
    แดง
    -0.29
     tantôt
    -0.28
    ρης
    -0.28
     dramatique
    -0.28
    POSITIVE LOGITS
    ⟬
    0.95
    GOTREF
    0.92
    
    0.90
    ########.
    0.90
    enderror
    0.85
    endpush
    0.84
    CppMethod
    0.81
     ब्रेकडाउन
    0.79
    tinyos
    0.79
    帖最后由
    0.79
    Activations Density 1.647%

    No Known Activations