© 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. 79214
    Prev
    Next
    INDEX
    Explanations

    topics related to community discussions and organizational transparency

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

    dialogue and question sessions

    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
     Ajoutez
    -0.31
     konkuren
    -0.31
     calorías
    -0.31
     sereia
    -0.31
     casada
    -0.31
     capitaine
    -0.30
     بيها
    -0.29
     województwie
    -0.28
     bumbu
    -0.28
    completed
    -0.28
    POSITIVE LOGITS
     Dialogue
    0.64
    Dialogue
    0.61
     dialogue
    0.61
     dialogues
    0.60
     Wikimedijinoj
    0.60
    VYMaps
    0.60
    ChrTalk
    0.59
    httphttps
    0.57
     Meet
    0.57
     DIALOG
    0.56
    Activations Density 0.018%

    No Known Activations