© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 24-GEMMASCOPE-RES-16K
    11. 11385
    Prev
    Next
    INDEX
    Explanations

    the word "which" and its variations in various contexts

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

    "which" often introducing a question

    np_acts-logits-general · gemini-2.0-flash

    which introduces clauses or questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/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.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    odyear
    -0.51
     Danilo
    -0.48
    moreland
    -0.47
    enal
    -0.46
    ada
    -0.46
    estimable
    -0.43
    ade
    -0.43
    tle
    -0.43
    ete
    -0.42
    ValueStyle
    -0.42
    POSITIVE LOGITS
     WHICH
    0.85
     Which
    0.84
    Which
    0.77
    which
    0.70
     Wich
    0.65
     which
    0.63
     والتي
    0.62
     wich
    0.59
     lesquels
    0.58
    igreja
    0.57
    Activations Density 0.222%

    No Known Activations