© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. DeepSeek-R1-Dist-Llama-8B
    5. 21-LLAMASCOPE-SLIMPJ-OPENR1-RES-32K
    7. 29831
    Prev
    Next
    INDEX
    Explanations

    the concept of a single boxed answer within a set of instructions

    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/L21R
    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.21.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ataka
    -0.06
    NU
    -0.06
     staging
    -0.06
    okes
    -0.06
     Split
    -0.06
    atak
    -0.06
     Spe
    -0.06
    uckle
    -0.05
    und
    -0.05
     all
    -0.05
    POSITIVE LOGITS
     boxed
    0.08
     box
    0.07
     single
    0.07
    within
    0.07
    eldo
    0.07
    usercontent
    0.06
     bold
    0.06
    boxed
    0.06
     boxes
    0.06
    ç½®
    0.06
    Activations Density 0.082%

    No Known Activations