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. 1608
    Prev
    Next
    INDEX
    Explanations

    instances of publication timestamps and details in text

    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
     hall
    -0.15
    867
    -0.15
    ÑĪÑĮ
    -0.14
    енÑĮ
    -0.14
    eding
    -0.14
     oneself
    -0.14
     NEC
    -0.14
    ÑĤе
    -0.14
    585
    -0.13
     Harr
    -0.13
    POSITIVE LOGITS
    achel
    0.19
    alah
    0.16
    dre
    0.16
    lash
    0.15
    rape
    0.15
    yg
    0.15
    .Atomic
    0.14
    itom
    0.14
    ãĤ¶
    0.14
    opher
    0.13
    Activations Density 0.045%

    No Known Activations