© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 16678
    Prev
    Next
    INDEX
    Explanations

    expressions of identity and personal experiences, especially relating to gender and privilege

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

    identifying traits and roles

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects short discourse/stance tokens and emphatic words — i.e., response words (yes/no/all), intensifiers and adverbs that convey emphasis or speaker stance.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     المعيارى
    -0.61
    hasMoreElements
    -0.59
    postsleuth
    -0.57
     <>",
    -0.56
     estekak
    -0.55
    DoubleQuotes
    -0.54
     OMITBAD
    -0.54
    DockStyle
    -0.54
     kaynağından
    -0.54
    Diweddarwch
    -0.54
    POSITIVE LOGITS
    Trait
    0.40
    Role
    0.33
    RoleId
    0.33
     trait
    0.32
     identity
    0.31
     Amu
    0.31
     vectorielles
    0.31
     Eres
    0.30
    instancetype
    0.30
     dumne
    0.30
    Activations Density 0.189%

    No Known Activations