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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. DeepSeek-R1-Dist-Llama-8B
    3. 21-LLAMASCOPE-SLIMPJ-OPENR1-RES-32K
    4. 29718
    Prev
    Next
    INDEX
    Explanations

    phrases indicating reflection on a mathematical problem-solving approach.

    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
    ÏĩÏİ
    -0.07
    andre
    -0.06
    ï¼Įé»ĺ认
    -0.06
    erece
    -0.06
     Hoch
    -0.06
    ivet
    -0.06
     vod
    -0.06
    yd
    -0.06
    oney
    -0.06
     StringType
    -0.06
    POSITIVE LOGITS
     nothing
    0.07
     unrelated
    0.07
     its
    0.06
    obe
    0.06
     Nothing
    0.06
     neither
    0.06
    ÅĻel
    0.06
     Twice
    0.06
     twice
    0.06
     Being
    0.06
    Activations Density 0.176%

    No Known Activations