© 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. 4-LLAMASCOPE-RES-32K
    11. 17135
    Prev
    Next
    INDEX
    Explanations

    the word "rather" and its variations, indicating a focus on comparisons or preferences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     addCriterion
    -0.18
    ãĥĸãĥª
    -0.16
    ZO
    -0.15
    ÑĥÑĢе
    -0.15
     then
    -0.15
     Vall
    -0.15
     intro
    -0.14
     dabei
    -0.14
    αÏģά
    -0.14
    Äĩe
    -0.14
    POSITIVE LOGITS
    retch
    0.16
     Bened
    0.15
    änger
    0.14
    ê°ģ
    0.13
    hire
    0.13
    çĶ·åŃIJ
    0.13
    hm
    0.13
    287
    0.13
    reich
    0.13
    íĬ¹
    0.13
    Activations Density 0.018%

    No Known Activations