© 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
    3. DeepSeek-R1-Dist-Llama-8B
    5. 17-LLAMASCOPE-SLIMPJ-OPENR1-RES-32K
    7. 23267
    Prev
    Next
    INDEX
    Explanations

    conditional phrases, especially hypothetical statements and reasons, commonly using words like "since", "as", "those", "would", "but", "because", "that".

    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/L17R
    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.17.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ovit
    -0.07
    annon
    -0.06
    enal
    -0.06
    agination
    -0.06
    otos
    -0.06
    ÑĢÑĥб
    -0.06
     Alb
    -0.05
    istrat
    -0.05
    HB
    -0.05
    .generated
    -0.05
    POSITIVE LOGITS
     would
    0.11
     risk
    0.10
    would
    0.09
     risks
    0.09
     Would
    0.09
     risking
    0.09
    Would
    0.09
     Ris
    0.08
    éĤ£æł·
    0.08
     Doing
    0.08
    Activations Density 0.082%

    No Known Activations