© 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. 12-RES-MATRYOSHKA-DC
    7. 356
    Prev
    Next
    INDEX
    Explanations

    phrases related to accusations of dishonesty or impropriety

    oai_token-act-pair · gpt-4o-mini
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    chanind/pile-uncopyrighted-gemma-1024-abbrv-1B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    featureID
    -0.55
    rens
    -0.53
    empre
    -0.53
    ン
    -0.52
     vieles
    -0.51
    inghouse
    -0.50
     veng
    -0.50
    queryInterface
    -0.49
     clearfix
    -0.49
    ScrollPane
    -0.47
    POSITIVE LOGITS
     people
    0.86
     fellow
    0.76
     whoever
    0.76
    onAttach
    0.75
     whome
    0.73
     me
    0.69
    ผู้
    0.68
     us
    0.68
     anyone
    0.67
    raszam
    0.66
    Activations Density 4.072%

    No Known Activations