© 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. DeepSeek-R1-Dist-Llama-8B
    5. 17-LLAMASCOPE-SLIMPJ-OPENR1-RES-32K
    7. 23391
    Prev
    Next
    INDEX
    Explanations

    a few words in math and logic problems, particularly "each", "doesn't" and words ending in "-celes"

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    fnlp/Llama-Scope-R1-Distill/400M-Slimpajama-400M-OpenR1-Math-220k/L17R
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Hzfinfdu/SlimPajama-3B and open-r1/OpenR1-Math-220k
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    /=
    -0.07
    oldem
    -0.07
     [~,
    -0.07
    alars
    -0.06
    imler
    -0.06
    à¸Ńà¸ļ
    -0.06
    gii
    -0.06
    Äĥng
    -0.06
     fitte
    -0.06
    важа
    -0.06
    POSITIVE LOGITS
    implify
    0.07
    [âĢ¦
    0.07
    ï¸ı
    0.07
     Abed
    0.06
    .initializeApp
    0.06
    gorm
    0.06
    MBER
    0.06
     dis
    0.06
    ité
    0.06
    <tbody
    0.06
    Activations Density 0.199%

    No Known Activations