© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-RES-MATRYOSHKA-DC
    7. 2926
    Prev
    Next
    INDEX
    Explanations

    words and phrases associated with lies and deception, including the phrase "fake news".

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.22.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.22.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
    Tikang
    -0.79
     '\\;'
    -0.70
     pinulongan
    -0.69
    DotNetBar
    -0.67
    MessageOf
    -0.63
    OGND
    -0.56
    reportWebVitals
    -0.56
     aapt
    -0.56
    ContextCompat
    -0.56
    UnknownFieldSet
    -0.55
    POSITIVE LOGITS
     fake
    2.69
    fake
    2.34
     Fake
    2.33
    Fake
    2.28
     faking
    2.16
     faked
    2.14
     pretend
    2.09
     pretending
    2.06
     fakes
    2.05
     phony
    2.05
    Activations Density 1.337%

    No Known Activations