© 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. 87287
    Prev
    Next
    INDEX
    Explanations

    punctuation, specifically commas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     Hosts
    -0.34
     Theater
    -0.33
    예
    -0.32
     HttpNotFound
    -0.32
     dry
    -0.31
     Theatre
    -0.31
     squ
    -0.31
    覗
    -0.31
     zak
    -0.31
    !!
    -0.30
    POSITIVE LOGITS
     autorytatywna
    0.92
    Geplaatst
    0.72
    OGND
    0.68
    Autoritní
    0.65
     يتيمه
    0.63
     informée
    0.62
     ſch
    0.57
     Wicidata
    0.55
     समीक्षाओं
    0.55
     ſont
    0.54
    Activations Density 0.000%

    No Known Activations