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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 36638
    Prev
    Next
    INDEX
    Explanations

    references to representation and diversity across various groups or categories

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

    groups represented by

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     gyhoeddwyd
    -0.37
     Púb
    -0.37
    ön
    -0.35
    SharedDtor
    -0.34
     Detta
    -0.32
    cyjnych
    -0.32
    synthetic
    -0.31
    şte
    -0.31
    iglio
    -0.31
     analytically
    -0.31
    POSITIVE LOGITS
     represented
    2.42
    represented
    2.16
     representation
    2.16
     Representation
    1.91
    representation
    1.90
     représent
    1.86
     REPRESENT
    1.84
     représenté
    1.83
    Representation
    1.83
     representado
    1.82
    Activations Density 0.093%

    No Known Activations