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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-2B
    3. 21-GEMMASCOPE-TRANSCODER-16K
    4. 6307
    Prev
    Next
    INDEX
    Explanations

    words related to home construction and housing security

    oai_token-act-pair · gemini-2.0-flash

    hardware components for doors

    np_token-act-pair-logits · gpt-4o-mini

    door handles

    np_max-act-logits · gemini-2.0-flash

    door Method used: 1 — because top activating tokens are door-related

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    queryInterface
    -0.71
     autorytatywna
    -0.60
     Gump
    -0.55
    rungsseite
    -0.51
    Autoritní
    -0.50
     AspNetCore
    -0.49
    mybatisplus
    -0.48
    IAH
    -0.47
     popoli
    -0.47
    帖最后由
    -0.47
    POSITIVE LOGITS
    ugc
    0.64
     hinges
    0.54
    ئية
    0.53
     handle
    0.51
     handles
    0.50
     lever
    0.49
     Allociné
    0.49
    Handle
    0.48
    handle
    0.48
    UseVisualStyle
    0.47
    Activations Density 1.180%

    No Known Activations