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. 8089
    Prev
    Next
    INDEX
    Explanations

    instances of the word "by" indicating authorship or origin

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

    No Comments

    Negative Logits
    <typeof
    -0.16
    noon
    -0.14
    à¥ĭà¤ķर
    -0.13
    undi
    -0.13
    staking
    -0.13
    ipient
    -0.13
     poil
    -0.13
    icient
    -0.13
     коÑĤ
    -0.13
    çon
    -0.13
    POSITIVE LOGITS
     virtue
    0.21
     means
    0.20
     chance
    0.19
     accident
    0.18
    -election
    0.18
     default
    0.17
    -pass
    0.17
     definition
    0.16
     way
    0.16
     design
    0.16
    Activations Density 0.146%

    No Known Activations