© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 63147
    Prev
    Next
    INDEX
    Explanations

    instances of the word "qui," indicating a focus on relative clauses or connections within sentences

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

    who or which in many languages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     christlichen
    -0.47
     enfans
    -0.43
    au
    -0.39
     bestuur
    -0.39
     Flucht
    -0.38
     Tinggi
    -0.38
     bénévol
    -0.37
     hvid
    -0.37
    assertRaises
    -0.36
     lèvres
    -0.36
    POSITIVE LOGITS
     والتي
    0.93
     والذي
    0.84
     которая
    0.81
     who
    0.79
     который
    0.79
     joka
    0.78
     który
    0.78
     która
    0.77
     WHICH
    0.76
     která
    0.76
    Activations Density 0.002%

    No Known Activations