© 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. 22035
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing the concept of "rightness" or correctness in various contexts

    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
    Falls
    -0.58
    ```
    -0.57
    gag
    -0.56
     glatt
    -0.56
    Curtis
    -0.56
     Grie
    -0.54
    seri
    -0.54
     practi
    -0.53
    rond
    -0.53
    shorts
    -0.51
    POSITIVE LOGITS
    脚注の使い方
    0.93
     IBOutlet
    0.77
     kasarigan
    0.76
    yntaxException
    0.71
    imanapun
    0.69
    NegativeButton
    0.68
     يتيمه
    0.66
     lenker
    0.65
    enumi
    0.63
    Hauptartikel
    0.62
    Activations Density 0.016%

    No Known Activations