© 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. 25-LLAMASCOPE-RES-32K
    11. 13633
    Prev
    Next
    INDEX
    Explanations

    phrases that express support or positive sentiment towards events or causes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rompt
    -0.16
    _AMD
    -0.15
    ico
    -0.15
    ycl
    -0.15
    th
    -0.15
     Firearms
    -0.14
    ___
    -0.14
    rej
    -0.14
    áo
    -0.14
    posites
    -0.13
    POSITIVE LOGITS
    æĺ¯ä¸ª
    0.30
    æĺ¯ä¸Ģ个
    0.26
     an
    0.22
    æĺ¯ä¸Ģ
    0.20
     SUCH
    0.18
     eine
    0.18
     such
    0.17
     a
    0.17
    een
    0.16
     má»Ļt
    0.15
    Activations Density 0.313%

    No Known Activations