Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 20-GEMMASCOPE-TRANSCODER-16K
    4. 16329
    Prev
    Next
    INDEX
    Explanations

    text about taking funds or guns away.

    oai_token-act-pair · gemini-2.0-flash

    taking away

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope/layer_20/width_16k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     BoxDecoration
    -0.50
     singularities
    -0.47
    ciale
    -0.46
    volen
    -0.46
    ScopeManager
    -0.45
    sidemargin
    -0.45
     urso
    -0.45
     eſt
    -0.45
     clôture
    -0.45
     Jefus
    -0.44
    POSITIVE LOGITS
     away
    1.43
     Away
    1.16
     AWAY
    1.10
    Away
    1.08
    taken
    1.08
     taken
    1.06
    away
    1.02
     snatched
    1.01
     Taken
    1.00
    Taken
    0.96
    Activations Density 1.029%

    No Known Activations