© 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. 30-LLAMASCOPE-RES-32K
    11. 24633
    Prev
    Next
    INDEX
    Explanations

    references to alcohol and its types in relation to hospitality settings

    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
     Cres
    -0.13
     Ballet
    -0.13
     Sew
    -0.13
     Bake
    -0.13
    feeding
    -0.13
    afone
    -0.13
     Tennis
    -0.13
     bake
    -0.12
     feeder
    -0.12
    \grid
    -0.12
    POSITIVE LOGITS
     drink
    0.56
     alcohol
    0.56
     drinking
    0.53
     drinkers
    0.52
     drinks
    0.52
     beer
    0.52
     alcoholic
    0.52
    éħĴ
    0.50
     booze
    0.49
     Alcohol
    0.48
    Activations Density 0.876%

    No Known Activations