© 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. 24-GEMMASCOPE-RES-16K
    6. 2001
    Prev
    Next
    INDEX
    Explanations

    references to programming languages and developer-related topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    "developer" or related terms

    np_acts-logits-general · gemini-2.0-flash

    software and developer roles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     larmes
    -0.42
    눔
    -0.39
     varones
    -0.39
    UNIDENTIFIED
    -0.39
     photographed
    -0.38
    UnusedPrivate
    -0.37
     disant
    -0.36
     fotografías
    -0.36
     women
    -0.36
     photographs
    -0.36
    POSITIVE LOGITS
     workflows
    0.69
     DevOps
    0.65
     usability
    0.63
     developer
    0.62
     workflow
    0.62
     developers
    0.62
     software
    0.59
     NSCoder
    0.54
     tooling
    0.54
     frameworks
    0.54
    Activations Density 1.529%

    No Known Activations