© 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. 31-GEMMASCOPE-RES-16K
    6. 4741
    Prev
    Next
    INDEX
    Explanations

    SQL keywords and clauses related to data manipulation and querying

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

    SQL syntax and database query structures.

    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
    RegressionTest
    -0.50
     &___
    -0.49
     surla
    -0.46
     EconPapers
    -0.43
     autorytatywna
    -0.43
    تقاوى
    -0.41
    Geplaatst
    -0.40
    odrome
    -0.37
     CanadaChoose
    -0.37
    +#+
    -0.36
    POSITIVE LOGITS
     join
    1.09
     joins
    1.01
    join
    0.98
     JOIN
    0.95
     Join
    0.93
     joining
    0.92
    Join
    0.90
     joined
    0.88
    JOIN
    0.87
     Joins
    0.86
    Activations Density 0.597%

    No Known Activations