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. 4471
    Prev
    Next
    INDEX
    Explanations

    references to community-based initiatives and approaches in various contexts

    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
    plus
    -0.14
    ovsky
    -0.14
     basically
    -0.14
    ãģķãģ¾
    -0.14
     jer
    -0.14
    oud
    -0.13
     presup
    -0.13
    upil
    -0.13
     dealt
    -0.13
    oky
    -0.13
    POSITIVE LOGITS
    ollen
    0.15
    μοί
    0.14
    ëŁ¼
    0.13
    ола
    0.13
    _PED
    0.13
     ped
    0.13
    pa
    0.13
     Likely
    0.13
    ushman
    0.13
    .localized
    0.13
    Activations Density 0.780%

    No Known Activations