© 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
    3. Gemma-2-2B
    5. 25-GEMMASCOPE-TRANSCODER-16K
    7. 9752
    Prev
    Next
    INDEX
    Explanations

    the words around 'a' followed by a term related to criminal activity

    oai_token-act-pair · gemini-2.0-flash

    crime and russia

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.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
    tvguidetime
    -0.71
     gynhyrchwyd
    -0.69
     estekak
    -0.68
    ItemBackground
    -0.65
     يتيمه
    -0.64
     متعلقه
    -0.64
    verwijspagina
    -0.63
    tagHelperRunner
    -0.60
    ImageContext
    -0.60
     CreateTagHelper
    -0.55
    POSITIVE LOGITS
     cref
    0.59
    Portale
    0.55
     fuego
    0.54
    tonode
    0.52
     públicas
    0.48
    CloseOperation
    0.47
    allerie
    0.47
    nec
    0.46
    PropTypes
    0.46
    ành
    0.46
    Activations Density 6.559%

    No Known Activations