© 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. 16-LLAMASCOPE-RES-32K
    11. 6476
    Prev
    Next
    INDEX
    Explanations

    occurrences of violence or weapon-related themes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Verfüg
    -0.15
     Gap
    -0.14
    Mul
    -0.14
    mul
    -0.14
    ureau
    -0.14
    erval
    -0.14
     gå
    -0.14
     pog
    -0.14
    alfa
    -0.13
    ittest
    -0.13
    POSITIVE LOGITS
    found
    0.23
     found
    0.23
     discovery
    0.22
    -find
    0.21
     finds
    0.21
     discovering
    0.20
     discovered
    0.20
     find
    0.19
     discovers
    0.19
     discoveries
    0.19
    Activations Density 0.121%

    No Known Activations