© 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. 21-LLAMASCOPE-RES-32K
    11. 15320
    Prev
    Next
    INDEX
    Explanations

    connections between causes and effects related to laws, negligence, and food choices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aira
    -0.16
    613
    -0.15
    aura
    -0.14
    ilst
    -0.14
    enate
    -0.13
    Ñŀ
    -0.13
    .wrapper
    -0.13
     Podesta
    -0.13
    ÏĦεÏģα
    -0.13
    istrovstvÃŃ
    -0.13
    POSITIVE LOGITS
     responsible
    0.63
     Responsible
    0.46
     responsable
    0.45
     contributing
    0.43
     contrib
    0.42
     contributor
    0.41
     contribute
    0.41
     responsibility
    0.40
    contrib
    0.40
    ponsible
    0.40
    Activations Density 0.226%

    No Known Activations