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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. DeepSeek-R1-Dist-Llama-8B
    3. 15-LLAMASCOPE-OPENR1-RES-32K
    4. 4
    Prev
    Next
    INDEX
    Explanations

    instances of numerical values and their relationships in mathematical problems

    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
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    (æ°´
    -0.07
    #ac
    -0.06
    ovit
    -0.06
    alles
    -0.06
    ãĤĥ
    -0.06
    _mux
    -0.06
    }elseif
    -0.06
     çł
    -0.06
    _TYP
    -0.06
    .avi
    -0.06
    POSITIVE LOGITS
    ses
    0.07
     either
    0.06
    itur
    0.06
    /dis
    0.06
    ARK
    0.06
    ldr
    0.06
     ú
    0.06
    aim
    0.06
     Third
    0.06
     auxiliary
    0.06
    Activations Density 0.013%

    No Known Activations