© 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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 97204
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate states of being or existence

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

    what is

    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
    posedge
    -0.81
    بوابة
    -0.65
    ьаж
    -0.63
     uș
    -0.62
     समीक्षाओं
    -0.61
    postIndex
    -0.59
    Portale
    -0.58
     ब्रेकडाउन
    -0.58
    MessageTagHelper
    -0.58
    AnchorStyles
    -0.55
    POSITIVE LOGITS
     quoi
    0.52
     what
    0.51
    是什么
    0.44
     nedir
    0.42
     hva
    0.41
    what
    0.40
     What
    0.39
     something
    0.39
    ArgsConstructor
    0.39
     WHAT
    0.38
    Activations Density 0.016%

    No Known Activations