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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 9-LLAMASCOPE-RES-32K
    11. 16144
    Prev
    Next
    INDEX
    Explanations

    phrases that involve controversial subjects and moral dilemmas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hone
    -0.16
    utin
    -0.16
    zend
    -0.14
    -Sah
    -0.14
    oret
    -0.13
    iž
    -0.13
    ourt
    -0.13
    ENDOR
    -0.13
    _primary
    -0.13
     Gund
    -0.13
    POSITIVE LOGITS
     behalf
    0.15
    erosis
    0.15
    unks
    0.14
    eza
    0.13
    ivé
    0.13
    rysler
    0.13
    áºŃt
    0.13
     Readonly
    0.13
     brand
    0.13
     mission
    0.13
    Activations Density 0.438%

    No Known Activations