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

    Neuronpedia

    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. 20501
    Prev
    Next
    INDEX
    Explanations

    references to specific animals and their characteristics or related concepts

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

    the word "old" followed by nouns, particularly when describing people, animals, or objects in a somewhat negative or derogatory context.

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

    mentions of age-related descriptions, especially the adjective indicating something is old preceding a noun.

    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
    231
    -0.15
    oux
    -0.15
     плÑı
    -0.14
    WithURL
    -0.14
    .descriptor
    -0.13
    Ñıви
    -0.13
    .follow
    -0.13
    ize
    -0.13
    wat
    -0.13
    stry
    -0.13
    POSITIVE LOGITS
    (old
    0.27
    -old
    0.27
    æŧ
    0.26
     old
    0.26
     OLD
    0.26
    old
    0.22
    OLD
    0.20
    Old
    0.20
    .old
    0.20
    /new
    0.19
    Activations Density 0.113%

    No Known Activations