© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 4700
    Prev
    Next
    INDEX
    Explanations

    phrases indicating contrast or exceptions

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
     NOT
    -0.22
     not
    -0.20
     nicht
    -0.19
     không
    -0.19
    ä¸įæĺ¯
    -0.19
     HOWEVER
    -0.18
     не
    -0.18
     niet
    -0.18
     ÙĦÙĬس
    -0.17
     नह
    -0.16
    POSITIVE LOGITS
     rather
    0.35
     Rather
    0.29
    Rather
    0.28
    rather
    0.28
     بÙĦÚ©Ùĩ
    0.20
     entire
    0.19
     actually
    0.19
     indeed
    0.18
     plutôt
    0.18
     mo
    0.18
    Activations Density 0.063%

    No Known Activations