© 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. 19602
    Prev
    Next
    INDEX
    Explanations

    strong assertions and claims that are hard to argue against or refute

    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
    IsMutable
    -0.68
    apimachinery
    -0.66
     defaultstate
    -0.66
    WriteTagHelper
    -0.61
    IndentedString
    -0.61
    jsxFileName
    -0.57
    istoitu
    -0.57
    imageshack
    -0.57
    :+:
    -0.57
     EconPapers
    -0.56
    POSITIVE LOGITS
    SequentialGroup
    0.61
    確かに
    0.61
    的确
    0.59
     legitimate
    0.58
     legit
    0.57
     legitimately
    0.56
     validity
    0.55
     INVENTION
    0.55
    确实
    0.54
     plausible
    0.54
    Activations Density 0.166%

    No Known Activations