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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 2302
    Prev
    Next
    INDEX
    Explanations

    names and personal pronouns in relational contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_15/width_16k/average_l0_131
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     gynhyrchwyd
    -0.52
     uLocal
    -0.48
     ceny
    -0.41
     wideo
    -0.41
    겜
    -0.40
    طلحات
    -0.39
    HostException
    -0.39
    IDAY
    -0.39
    libft
    -0.39
    CHEZ
    -0.38
    POSITIVE LOGITS
     Paglinawan
    0.62
    참고
    0.49
    ValueStyle
    0.47
     TAMBIÉN
    0.42
    DockStyle
    0.38
     himself
    0.38
    InsertCommand
    0.38
    principalColumn
    0.37
     actually
    0.37
     grandfather
    0.37
    Activations Density 0.498%

    No Known Activations