© 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. 27-LLAMASCOPE-RES-32K
    11. 5774
    Prev
    Next
    INDEX
    Explanations

    expressions related to limitation and lack of choice

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.27.hook_resid_post
    Hook Layer
    27
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     Yet
    -0.22
    Yet
    -0.20
    yet
    -0.19
     however
    -0.19
     yet
    -0.18
     Still
    -0.18
     allerdings
    -0.16
    placeholders
    -0.16
    pite
    -0.15
    ÏģοÏį
    -0.15
    POSITIVE LOGITS
     but
    0.46
     except
    0.41
    but
    0.38
     besides
    0.34
    except
    0.32
    ä½Ĩ
    0.32
     BUT
    0.32
     кÑĢоме
    0.29
     Except
    0.28
    -but
    0.28
    Activations Density 0.088%

    No Known Activations