© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 81012
    Prev
    Next
    INDEX
    Explanations

    instances of the word "carve" and its variations related to creating or shaping objects

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

    carving, chisel, etching

    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
     ControllerBase
    -0.48
     Permanente
    -0.47
     sys
    -0.47
     للمعارف
    -0.45
     expect
    -0.45
     Sys
    -0.44
     Gea
    -0.42
     informée
    -0.41
    Plus
    -0.41
    SuccessListener
    -0.41
    POSITIVE LOGITS
     carving
    1.02
    carved
    0.96
     carved
    0.93
     carve
    0.92
     carvings
    0.79
     escultura
    0.65
     sculpture
    0.60
     cutting
    0.60
    Cutting
    0.59
     sculpted
    0.59
    Activations Density 0.008%

    No Known Activations