© 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. 38-GEMMASCOPE-RES-16K
    6. 10021
    Prev
    Next
    INDEX
    Explanations

    terms related to architectural features and construction

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    +#+#
    -0.58
     Administrativna
    -0.51
    prepareStatement
    -0.48
    tomans
    -0.48
    autique
    -0.46
    HtmlAttribute
    -0.45
    tagHelperRunner
    -0.44
    Portály
    -0.44
    ponses
    -0.43
     createSlice
    -0.43
    POSITIVE LOGITS
     roof
    2.44
     Roof
    2.03
    Roof
    1.98
    roof
    1.97
     roofs
    1.93
     ROOF
    1.88
     roofing
    1.67
     rooftop
    1.60
     rooftops
    1.48
    屋根
    1.35
    Activations Density 0.271%

    No Known Activations