© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 5810
    Prev
    Next
    INDEX
    Explanations

    words and phrases that express something bad is going to happen, especially in the context of technology, legal issues or personal relationships.

    oai_token-act-pair · gemini-2.0-flash

    avoiding negative consequences

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     disfrut
    -0.50
    Catégorie
    -0.47
    ilova
    -0.47
    ভা
    -0.45
     casque
    -0.44
    appspot
    -0.44
     مشين
    -0.44
     provato
    -0.41
    foro
    -0.41
    הערות
    -0.41
    POSITIVE LOGITS
    发表于
    0.68
    AsUp
    0.63
     nahilalakip
    0.63
    PreferredItem
    0.61
     चीज़ों
    0.59
    LookAnd
    0.59
    NameInMap
    0.58
    oneofs
    0.58
    stdc
    0.55
     Réponses
    0.55
    Activations Density 5.818%

    No Known Activations