© 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. 2-LLAMASCOPE-RES-32K
    11. 8171
    Prev
    Next
    INDEX
    Explanations

    verbs and their various forms, particularly those ending in 'ing' and 'ed'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    i
    -0.18
    ÛĮ
    -0.16
    iare
    -0.15
    .usage
    -0.15
    ing
    -0.14
    ision
    -0.14
    olucion
    -0.14
    ä¸ĢçĤ¹
    -0.13
    اÙĨ
    -0.13
    sak
    -0.13
    POSITIVE LOGITS
    erg
    0.15
    0.15
    ero
    0.13
    jee
    0.13
    hire
    0.13
    attice
    0.13
     Kerr
    0.13
    his
    0.13
    erc
    0.13
    ipp
    0.13
    Activations Density 0.253%

    No Known Activations