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. 4498
    Prev
    Next
    INDEX
    Explanations

    references to third-party services and their impacts 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
    ondo
    -0.15
    /preferences
    -0.15
     pesso
    -0.15
    /alert
    -0.14
    ND
    -0.14
    urry
    -0.13
    коÑĤ
    -0.13
    Äįan
    -0.13
    _nd
    -0.13
    otti
    -0.13
    POSITIVE LOGITS
     third
    0.18
    /internal
    0.17
     external
    0.17
    vise
    0.17
     outside
    0.16
    idget
    0.15
    isma
    0.15
     第ä¸ī
    0.14
    第ä¸ī
    0.14
    afil
    0.14
    Activations Density 0.045%

    No Known Activations