© 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 - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 13266
    Prev
    Next
    INDEX
    Explanations

    field or column names in structured data formats, particularly those related to database queries or JSON structures.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @myinzi123

    identifiers and attributes related to data structures or objects in programming

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

    Code field names and datatypes

    np_acts-logits-general · gemini-2.0-flash

    labels and identifiers

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    -0.85
    AndEndTag
    -0.63
    enumii
    -0.62
    EndContext
    -0.58
    DockStyle
    -0.57
    ConstraintMaker
    -0.56
     Chwiliwch
    -0.56
     MessageBoxIcon
    -0.54
    Espèce
    -0.53
    ="@+
    -0.53
    POSITIVE LOGITS
    veien
    0.44
    NSUInteger
    0.43
    formación
    0.42
     Gerichts
    0.40
     plateado
    0.39
    utscher
    0.39
     representation
    0.39
     électronique
    0.38
     ortop
    0.38
     enfermed
    0.38
    Activations Density 0.335%

    No Known Activations