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-SLIMPJ-RES-32K
    4. 4458
    Prev
    Next
    INDEX
    Explanations

    concepts related to broad themes, regulations, and domestic items

    oai_token-act-pair · gpt-4o-mini
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    fnlp/Llama-Scope-R1-Distill/800M-Slimpajama-0-OpenR1-Math-220k/L15R
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Hzfinfdu/SlimPajama-3B
    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
    elay
    -0.17
    laÄį
    -0.17
    abei
    -0.16
    lesc
    -0.16
    epam
    -0.15
    ipop
    -0.15
    isay
    -0.15
    alic
    -0.14
    EFAULT
    -0.14
    erli
    -0.14
    POSITIVE LOGITS
    eck
    0.15
     
    0.15
     quoted
    0.14
     Wend
    0.14
     Geh
    0.13
     retry
    0.13
    ()
    0.13
    ubi
    0.13
     Proud
    0.12
    ate
    0.12
    Activations Density 0.237%

    No Known Activations