© 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. 24-RES-MATRYOSHKA-DC
    7. 20
    Prev
    Next
    INDEX
    Explanations

    first-person narrative and conversational language

    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.24.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.24.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
     quæ
    -1.24
     auffi
    -1.19
     ſtate
    -1.19
     avoient
    -1.17
    ſelf
    -1.14
     ciasc
    -1.13
     ainfi
    -1.12
     stället
    -1.10
     asimismo
    -1.09
     iſt
    -1.08
    POSITIVE LOGITS
     kinda
    1.29
     really
    1.24
     crappy
    1.24
     REALLY
    1.20
     pretty
    1.19
     stuff
    1.17
     crap
    1.16
     crazy
    1.12
     my
    1.09
     nice
    1.09
    Activations Density 11.370%

    No Known Activations