Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 67001
    Prev
    Next
    INDEX
    Explanations

    instances of the word "app."

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

    app followed by icator or ium

    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
    :✨
    -0.56
     arşivlendi
    -0.49
    IVEREF
    -0.44
     Italijanski
    -0.42
    ₤
    -0.41
    帖最后由
    -0.40
     tocado
    -0.40
    fromnode
    -0.39
    endregion
    -0.38
    พาะ
    -0.38
    POSITIVE LOGITS
     houſe
    0.53
    ⪜
    0.52
    textra
    0.52
     Houſe
    0.52
     observable
    0.52
     pleaſure
    0.50
    kapp
    0.50
     ſtate
    0.50
     LST
    0.49
     hunne
    0.48
    Activations Density 0.000%

    No Known Activations