© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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 - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 102930
    Prev
    Next
    INDEX
    Explanations

    words and expressions reflecting emotional sentiments and connections related to personal experiences and relationships

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

    Small connecting elements in text, primarily punctuation marks (apostrophes, commas, hyphens) and occasionally short connector words, often appearing at transition points in sentences or in technical/digital contexts.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tagHelperRunner
    -0.43
     plenty
    -0.37
    yntaxException
    -0.36
     فريبيس
    -0.35
    Plenty
    -0.35
     виправивши
    -0.33
    AddTagHelper
    -0.33
     instances
    -0.33
    SerializedSize
    -0.32
    Достоинства
    -0.31
    POSITIVE LOGITS
     ब्रेकडाउन
    0.65
     initComponents
    0.52
    Personensuche
    0.51
     ostavi
    0.50
    ORPG
    0.49
     Paglinawan
    0.48
    supset
    0.47
    WireFormatLite
    0.47
    +#+#
    0.47
     autorytatywna
    0.46
    Activations Density 0.043%

    No Known Activations