© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 20-LLAMASCOPE-RES-32K
    11. 15412
    Prev
    Next
    INDEX
    Explanations

    suffixes and prefixes indicating status, condition, or property

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    halt
    -0.22
    ãģĦãĤĭ
    -0.20
    rng
    -0.18
    ãģĦãģŁ
    -0.18
    à¥įद
    -0.18
    rms
    -0.18
    hook
    -0.17
    rut
    -0.17
    hind
    -0.17
    à¥įड
    -0.17
    POSITIVE LOGITS
    ech
    0.29
    issue
    0.23
    ools
    0.23
    egra
    0.23
    egrity
    0.23
    eam
    0.22
    ãĤ¥
    0.22
    undra
    0.22
    ypes
    0.21
    roph
    0.21
    Activations Density 0.429%

    No Known Activations