© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. DeepSeek-R1-Dist-Llama-8B
    3. 21-LLAMASCOPE-SLIMPJ-OPENR1-RES-32K
    4. 30003
    Prev
    Next
    INDEX
    Explanations

    the characters "://", often found in URLs, and also it seemingly dislikes the combination of "not necessarily have" (though it's very possible this is noise)

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    fnlp/Llama-Scope-R1-Distill/400M-Slimpajama-400M-OpenR1-Math-220k/L21R
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Hzfinfdu/SlimPajama-3B and open-r1/OpenR1-Math-220k
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .LookAndFeel
    -0.09
    ÅĻÃŃzenÃŃ
    -0.07
    IFI
    -0.07
     mastur
    -0.06
    oldur
    -0.06
    itler
    -0.06
    adele
    -0.06
    Äł
    -0.06
    ÅĻÃŃd
    -0.06
    jas
    -0.06
    POSITIVE LOGITS
    enberg
    0.07
     initial
    0.06
     almost
    0.06
    ’s
    0.06
     double
    0.06
     recess
    0.06
     quot
    0.06
    па
    0.06
    kers
    0.06
     Fre
    0.06
    Activations Density 0.183%

    No Known Activations