© 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. 12329
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal sentences and punishments

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

    legal and criminal justice proceedings involving sentences, crimes, arrests, and court actions.

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

    text describing political conflicts, legal proceedings, or organized groups involved in civil unrest or criminal activities.

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

    references to state security and criminal justice actions—such as policing, arrests, charges, sentencing, and control of public gatherings—often in political or conflict 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
    紐
    -0.46
    Rptr
    -0.45
     noDo
    -0.44
    California
    -0.44
     California
    -0.42
     Wm
    -0.42
     Algon
    -0.42
     Америка
    -0.41
     typelib
    -0.41
     Hispanic
    -0.40
    POSITIVE LOGITS
     EconPapers
    0.76
     nationalist
    0.76
     Islamist
    0.71
     Nationalist
    0.69
     nationalists
    0.68
     impunity
    0.65
    authorised
    0.59
     Officially
    0.59
     envisaged
    0.58
     reformas
    0.57
    Activations Density 1.129%

    No Known Activations