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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. DeepSeek-R1-Dist-Llama-8B
    3. 15-LLAMASCOPE-OPENR1-RES-32K
    4. 11
    Prev
    Next
    INDEX
    Explanations

    mathematical equations or calculations

    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
    inger
    -0.07
     nonlinear
    -0.06
    wh
    -0.06
    eno
    -0.06
    arg
    -0.06
    act
    -0.06
    pg
    -0.06
    ai
    -0.06
    lev
    -0.06
    seg
    -0.06
    POSITIVE LOGITS
    IDGET
    0.07
    upo
    0.07
    ronics
    0.07
    hetto
    0.07
    pheres
    0.07
     addCriterion
    0.06
    rubu
    0.06
    elper
    0.06
    ÙħÙĪÙĦ
    0.06
    /Dk
    0.06
    Activations Density 0.118%

    No Known Activations