© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 15-LLAMASCOPE-RES-32K
    11. 11909
    Prev
    Next
    INDEX
    Explanations

    phrases indicating central importance or foundational elements in a context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    the phrase "at the heart of" or similar expressions indicating something central or core to a topic.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    phrases that describe the central or essential part of something, especially constructions like “at the … of …” indicating the core of a subject.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    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
    loh
    -0.17
    edom
    -0.16
    vince
    -0.15
    shima
    -0.15
    gan
    -0.15
     Richardson
    -0.15
    _mgmt
    -0.14
    ádu
    -0.14
    -prev
    -0.13
    anth
    -0.13
    POSITIVE LOGITS
    /core
    0.22
     core
    0.21
    core
    0.19
    (core
    0.18
    æł¸å¿ĥ
    0.17
    _core
    0.16
     foundation
    0.16
    -core
    0.16
     cores
    0.16
    .shtml
    0.16
    Activations Density 0.102%

    No Known Activations