© 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 - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 15402
    Prev
    Next
    INDEX
    Explanations

    direct quotes or dialogue in the text

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

    direct quotes or speech that convey emotional statements or personal declarations.

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

    quotations expressing personal commitment, determination, or strong opinions, particularly in political or emotional contexts.

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

    directly quoted speech enclosed in quotation marks, especially statements from people in news or political contexts.

    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
     Survi
    -0.43
    XmlRootElement
    -0.43
     definitely
    -0.41
     really
    -0.40
    definitely
    -0.40
     Lam
    -0.39
     Ritual
    -0.38
    brücken
    -0.38
     Really
    -0.36
     Bech
    -0.36
    POSITIVE LOGITS
    :✨
    0.58
    SequentialGroup
    0.57
    DockStyle
    0.53
    存于互联网档案馆
    0.51
     proposito
    0.49
     circunstancia
    0.47
    ministrazione
    0.47
    EndContext
    0.47
     كومونز
    0.47
    Folks
    0.46
    Activations Density 0.223%

    No Known Activations