© 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. 20210
    Prev
    Next
    INDEX
    Explanations

    terms related to unlocking, unemployment, and uniformity

    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
    adera
    -0.17
    izr
    -0.16
    äºİ
    -0.16
    oner
    -0.15
    activ
    -0.15
     unanim
    -0.15
    /use
    -0.14
    ä»ĺãģij
    -0.14
    	↵↵
    -0.14
    out
    -0.14
    POSITIVE LOGITS
    (Un
    0.21
    anness
    0.20
    /un
    0.19
    ión
    0.18
    uguay
    0.17
    lug
    0.17
    esco
    0.17
    iverse
    0.17
    ikit
    0.17
    ertainty
    0.17
    Activations Density 0.078%

    No Known Activations