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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 7691
    Prev
    Next
    INDEX
    Explanations

    data types and memory management functions in programming contexts

    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
    []
    -0.35
     [...]
    -0.33
     being
    -0.33
     strlen
    -0.32
     paramInt
    -0.32
    MoveNext
    -0.31
    ...]
    -0.31
    being
    -0.31
     beberapa
    -0.31
     œuvres
    -0.30
    POSITIVE LOGITS
     const
    0.86
    jsxFileName
    0.75
     مرئيه
    0.70
    GenerationType
    0.66
    Personendaten
    0.65
    fjspx
    0.64
    makeConstraints
    0.63
    
    0.61
     Normdatei
    0.60
    **
    0.59
    Activations Density 0.344%

    No Known Activations