© 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. 59022
    Prev
    Next
    INDEX
    Explanations

    data structures and their manipulation in programming code

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

    list and dictionary manipulations

    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
    endfor
    -0.63
     endfor
    -0.60
    onAttach
    -0.57
     ब्रेकडाउन
    -0.57
     Chwiliwch
    -0.56
    [@BOS@]
    -0.54
    SharedDtor
    -0.54
    <unused51>
    -0.54
    <unused41>
    -0.54
    <unused16>
    -0.54
    POSITIVE LOGITS
     entière
    0.42
    라는
    0.36
     graphique
    0.36
     collection
    0.34
     chrétien
    0.33
     écrite
    0.33
    谱
    0.33
     voici
    0.33
     recueil
    0.33
    kowa
    0.31
    Activations Density 0.013%

    No Known Activations