© 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. 6221
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "secrets."

    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
    Negative Logits
     Koss
    -0.76
    NameInMap
    -0.75
    __*/
    -0.75
    ✨:
    -0.74
     становника
    -0.74
     LIF
    -0.74
    MigrationBuilder
    -0.72
    romyalgia
    -0.71
    xffffffff
    -0.71
    harusnya
    -0.70
    POSITIVE LOGITS
     Secret
    1.63
     SECRET
    1.63
     Secrets
    1.63
    Secret
    1.60
     secrets
    1.58
    Secrets
    1.48
     secret
    1.47
    secret
    1.45
    secrets
    1.38
    SECRET
    1.38
    Activations Density 0.055%

    No Known Activations