© 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. 30-LLAMASCOPE-RES-32K
    11. 12143
    Prev
    Next
    INDEX
    Explanations

    references to comedy and humor in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uries
    -0.16
     prelim
    -0.13
    KANJI
    -0.13
     heyec
    -0.13
    PullParser
    -0.13
     ëĤĺê°Ģ
    -0.13
    alker
    -0.13
    icas
    -0.13
     Hakk
    -0.12
    erah
    -0.12
    POSITIVE LOGITS
     comedy
    0.47
     humor
    0.46
     humour
    0.43
     comedic
    0.41
     laughs
    0.40
     satire
    0.40
     Comedy
    0.39
     humorous
    0.39
     laughter
    0.38
     jokes
    0.38
    Activations Density 1.159%

    No Known Activations