© 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. 20-GEMMASCOPE-RES-16K
    11. 5595
    Prev
    Next
    INDEX
    Explanations

    references to racial identity and political affiliations

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

    details about racial or ethnic identity and classification.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    phrases that assert or classify a person’s identity, status, or group affiliation (e.g., relationships, titles/roles, party/tribal membership, ethnicity/ancestry, sexual orientation).

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    phrases that identify, describe, or claim someone's race, ethnicity, or social status.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    +#+#
    -0.54
     BoxFit
    -0.53
     PACIFIC
    -0.48
    太平洋
    -0.48
    PutMapping
    -0.47
    OLEAN
    -0.47
    揄
    -0.46
     mechan
    -0.45
     destru
    -0.45
    lwjgl
    -0.45
    POSITIVE LOGITS
     possu
    0.49
    CppMethod
    0.44
     identitas
    0.43
     otomatig
    0.43
     Мексичка
    0.41
     possui
    0.41
    //
    0.40
     posesión
    0.40
     identidad
    0.38
     affiliations
    0.38
    Activations Density 0.388%

    No Known Activations