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. 4502
    Prev
    Next
    INDEX
    Explanations

    concepts related to persistence and ongoing efforts

    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
    YE
    -0.15
    Reuse
    -0.15
     Trey
    -0.15
    िह
    -0.15
    幸
    -0.15
    olini
    -0.14
    tre
    -0.14
    adel
    -0.14
    239
    -0.14
     Tre
    -0.14
    POSITIVE LOGITS
     seemingly
    0.21
     ê³ĦìĨį
    0.18
     continue
    0.17
    amus
    0.17
     continues
    0.17
     never
    0.16
     endless
    0.16
    ä¸įæĸŃ
    0.16
    enheim
    0.15
    contin
    0.15
    Activations Density 0.092%

    No Known Activations