© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 13761
    Prev
    Next
    INDEX
    Explanations

    legal terminology and concepts surrounding court rulings and procedures

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

    legal language expressing judicial deference to lower court decisions.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou

    legal or technical language indicating that a decision or ruling will not be overturned or reversed except under specific stringent conditions.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    language describing appellate standards of review, especially deferential statements about upholding or reversing trial court rulings and judgments.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    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
    RTCK
    -0.37
    лтемелер
    -0.31
    تقاوى
    -0.31
    verständlich
    -0.30
     ویکی‌پدیای
    -0.30
     فہرست
    -0.30
     eventual
    -0.29
    ⊝
    -0.29
     ternyata
    -0.29
     vindicated
    -0.28
    POSITIVE LOGITS
     unless
    3.34
    unless
    2.98
     Unless
    2.97
    Unless
    2.94
    除非
    2.44
     kecuali
    1.86
     sauf
    1.48
     except
    1.42
    except
    1.27
     Except
    1.19
    Activations Density 0.841%

    No Known Activations