Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. DeepSeek-R1-Dist-Llama-8B
    3. 15-LLAMASCOPE-OPENR1-RES-32K
    4. 76
    Prev
    Next
    INDEX
    Explanations

    mathematical concepts and relationships related to geometry and equations

    oai_token-act-pair · gpt-4o-mini
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    fnlp/Llama-Scope-R1-Distill/0-Slimpajama-800M-OpenR1-Math-220k/L15R
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    open-r1/OpenR1-Math-220k
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MUX
    -0.06
    hall
    -0.06
    èįī
    -0.06
     paz
    -0.06
    arta
    -0.06
    CAD
    -0.06
    Hum
    -0.06
    Demand
    -0.06
    uro
    -0.06
    orgen
    -0.06
    POSITIVE LOGITS
    untime
    0.08
    -runtime
    0.07
     runtime
    0.06
    ãĥ¥
    0.06
    iore
    0.06
    онÑĮ
    0.06
    è¿IJè¡Į
    0.06
     Dup
    0.06
    runtime
    0.06
     Howe
    0.06
    Activations Density 0.368%

    No Known Activations