© 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. 6-LLAMASCOPE-RES-32K
    11. 20081
    Prev
    Next
    INDEX
    Explanations

    terms related to danger and risky situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.6.hook_resid_post
    Hook Layer
    6
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    enance
    -0.18
    tle
    -0.15
     noqa
    -0.15
    CALE
    -0.15
    å±
    -0.15
    tings
    -0.14
    bred
    -0.14
    anity
    -0.14
    ìĦ¸ëĮĢ
    -0.14
    ç§
    -0.14
    POSITIVE LOGITS
    -danger
    0.22
     dangerous
    0.19
    ously
    0.18
    danger
    0.17
     dangers
    0.16
     Dangerous
    0.15
     danger
    0.15
     unsafe
    0.15
    essler
    0.14
    ous
    0.14
    Activations Density 0.020%

    No Known Activations