© 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. 23432
    Prev
    Next
    INDEX
    Explanations

    mentions of security vulnerabilities and detection evasion techniques

    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
    ausal
    -0.16
    etsy
    -0.15
    WISE
    -0.15
    ÑĢава
    -0.15
    баÑģ
    -0.15
    reement
    -0.15
    BJECT
    -0.14
    templ
    -0.14
    Thunk
    -0.14
    KER
    -0.14
    POSITIVE LOGITS
     Mol
    0.17
    vak
    0.16
     Frag
    0.15
     patterns
    0.15
    ignet
    0.15
    877
    0.15
     round
    0.14
     move
    0.14
    mol
    0.14
    IQ
    0.14
    Activations Density 0.290%

    No Known Activations