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. 36-GEMMASCOPE-RES-16K
    6. 16345
    Prev
    Next
    INDEX
    Explanations

    numerical values and comparisons in a scientific context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.36.hook_resid_post
    Hook Layer
    36
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    +#+
    -0.42
    migrationBuilder
    -0.42
    haikusbot
    -0.31
     wireType
    -0.30
    Brainz
    -0.28
    XmlRootElement
    -0.28
    Rptr
    -0.28
     ці
    -0.27
     ga
    -0.27
     Wrapper
    -0.27
    POSITIVE LOGITS
    tagHelperRunner
    0.65
    évaluateur
    0.59
    出版年
    0.57
     Grüsse
    0.57
    ніципалі
    0.56
    rsiniz
    0.55
    Życiorys
    0.53
     appelez
    0.53
    weightedMode
    0.53
     otomatig
    0.52
    Activations Density 0.611%

    No Known Activations