© 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
    2. Gemma-2-2B
    3. 24-GEMMASCOPE-TRANSCODER-16K
    4. 3962
    Prev
    Next
    INDEX
    Explanations

    groups of words expressing contrasting/adversarial ideas

    oai_token-act-pair · gemini-2.0-flash

    political/social commentary

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.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
    envolvimento
    -0.55
    日閲覧
    -0.50
    addPreferredGap
    -0.46
    olone
    -0.43
     initState
    -0.43
     authService
    -0.42
    ando
    -0.41
    andas
    -0.41
    oraz
    -0.41
    bonucle
    -0.41
    POSITIVE LOGITS
     myſelf
    0.58
     Италијани
    0.57
     purpoſe
    0.57
     houſe
    0.56
     дописавши
    0.56
     himſelf
    0.54
    حياتها
    0.53
     للاسماء
    0.53
    atchewan
    0.53
    Билгалдахарш
    0.53
    Activations Density 3.239%

    No Known Activations