© 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. 31-GEMMASCOPE-RES-16K
    6. 3599
    Prev
    Next
    INDEX
    Explanations

    references to evidence and testimonies linked to a legal case or investigation

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

    periods and punctuation marks that conclude sentences or clauses.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/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.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DotNetBar
    -0.59
    WaitGroup
    -0.57
    XmlAccessorType
    -0.56
     utafitiHapana
    -0.51
    ValueStyle
    -0.51
    ContentAsync
    -0.48
     WIC
    -0.47
    олові
    -0.46
     oprot
    -0.45
     betweenstory
    -0.45
    POSITIVE LOGITS
    《
    0.56
    ,《
    0.44
    <code>
    0.41
    ...@
    0.41
     sogen
    0.40
     kasarigan
    0.37
     sogenannte
    0.37
     sogenannten
    0.37
    》、《
    0.36
     satunya
    0.36
    Activations Density 1.930%

    No Known Activations