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. 32
    Prev
    Next
    INDEX
    Explanations

    conditional statements and their implications in mathematical reasoning

    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
     approximately
    -0.08
     PoÄįet
    -0.07
     Approximately
    -0.07
    .browser
    -0.06
    .scalablytyped
    -0.06
    å©Ĩ
    -0.06
    alent
    -0.06
     Affero
    -0.06
    OMPI
    -0.06
    ga
    -0.06
    POSITIVE LOGITS
     may
    0.09
    may
    0.07
    æŁIJ
    0.07
    šak
    0.07
    iale
    0.07
     perfectly
    0.07
     might
    0.07
     maybe
    0.07
     ÙħØ«ÙĦا
    0.07
     someone
    0.06
    Activations Density 0.066%

    No Known Activations