© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 14970
    Prev
    Next
    INDEX
    Explanations

    positive descriptive words when talking about a place

    oai_token-act-pair · gemini-2.0-flash

    words related to timing or events occurring later

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

    nightlife and entertainment

    np_max-act-logits · gemini-2.0-flash
    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
     ujednoznacz
    -0.69
    ILayout
    -0.60
    KommentareTeilen
    -0.60
    migrationBuilder
    -0.58
     propOrder
    -0.58
    ígenes
    -0.57
     chêne
    -0.57
    μιουργ
    -0.56
     gyhoeddwyd
    -0.55
    DebuggerNonUser
    -0.55
    POSITIVE LOGITS
     late
    0.75
     Late
    0.66
    Late
    0.61
    late
    0.60
     after
    0.59
     post
    0.53
    folge
    0.51
    anut
    0.50
    LEADING
    0.49
     follow
    0.47
    Activations Density 1.644%

    No Known Activations