© 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. 19992
    Prev
    Next
    INDEX
    Explanations

    key terms related to governance, societal responsibility, and economic impacts

    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
    éĩĮçļĦ
    -0.20
    far
    -0.17
    endi
    -0.16
    _far
    -0.16
    assin
    -0.16
    Away
    -0.15
    ä¼ij
    -0.15
    Into
    -0.15
    isoft
    -0.15
    اÙĨات
    -0.14
    POSITIVE LOGITS
     above
    0.30
     below
    0.23
    _above
    0.22
    above
    0.20
     Above
    0.20
    Above
    0.19
    _below
    0.18
     abaixo
    0.16
     directly
    0.16
    539
    0.15
    Activations Density 0.145%

    No Known Activations