© 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. 12-RES-MATRYOSHKA-DC
    7. 26662
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate instructions or reminders for actions to take

    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
     Мексичка
    -0.92
    cipar
    -0.76
     Sce
    -0.75
     Efq
    -0.75
     [?]
    -0.69
    ſelf
    -0.68
     '<?
    -0.66
     Houſe
    -0.66
     parlour
    -0.65
    ubli
    -0.65
    POSITIVE LOGITS
    бяза
    0.72
     обязательно
    0.71
     include
    0.69
     properly
    0.67
     обов
    0.61
     incluir
    0.60
    ستین
    0.60
    记得
    0.59
    getDrawable
    0.59
    还得
    0.59
    Activations Density 0.009%

    No Known Activations