© 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. 30-LLAMASCOPE-RES-32K
    11. 27352
    Prev
    Next
    INDEX
    Explanations

    the presence of the substring 'st' in various forms, often related to 'stoves' or 'stumps.'

    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
     Hammer
    -0.17
    osen
    -0.16
     thumb
    -0.15
    lbrace
    -0.14
    .scalablytyped
    -0.14
     uncomment
    -0.14
    ugin
    -0.14
    851
    -0.14
    849
    -0.14
    ITIVE
    -0.14
    POSITIVE LOGITS
    enc
    0.28
    ucc
    0.27
    urdy
    0.23
    airs
    0.23
    ained
    0.22
    air
    0.22
    ainless
    0.22
    encil
    0.22
    udded
    0.22
    urd
    0.22
    Activations Density 0.018%

    No Known Activations