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. 85891
    Prev
    Next
    INDEX
    Explanations

    references to familial relationships and lineage, particularly focusing on sons and children

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

    father or parent figures

    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
    tagHelperRunner
    -0.60
    NameInMap
    -0.46
    Datuak
    -0.44
    OGND
    -0.42
    ագրություններ
    -0.41
     trafi
    -0.40
    下载附件
    -0.40
     conclusione
    -0.40
    DeleteBehavior
    -0.40
    ymce
    -0.40
    POSITIVE LOGITS
     father
    0.59
     parent
    0.56
     parents
    0.56
     Father
    0.52
     mother
    0.52
     sons
    0.49
     fathers
    0.49
    setParent
    0.47
     FATHER
    0.47
     parental
    0.47
    Activations Density 0.042%

    No Known Activations