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. 1438
    Prev
    Next
    INDEX
    Explanations

    specific details or questions regarding procedures and requirements, particularly related to voting and registration

    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
    283
    -0.15
    IFICATIONS
    -0.14
    887
    -0.14
    teri
    -0.14
    pNet
    -0.14
    888
    -0.14
    lage
    -0.14
    mpar
    -0.14
    BUM
    -0.14
    IRM
    -0.13
    POSITIVE LOGITS
     longest
    0.27
     longer
    0.26
    long
    0.25
     dÃłi
    0.24
     length
    0.24
     long
    0.23
     Longer
    0.23
    -long
    0.21
    length
    0.21
     lengths
    0.21
    Activations Density 0.088%

    No Known Activations