© 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 EvalsExportsAPI 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. 20915
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate personal struggles or conflicts

    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
    onya
    -0.16
    unlikely
    -0.15
    kker
    -0.14
     myself
    -0.14
    éĤ£ä¹Ī
    -0.14
    ãģ¾ãģļ
    -0.14
    vp
    -0.14
    ogue
    -0.13
    ÑĢеÑģ
    -0.13
     had
    -0.13
    POSITIVE LOGITS
     lately
    0.39
     recently
    0.38
     sometimes
    0.35
     occasionally
    0.35
     whenever
    0.34
     Recently
    0.34
    Recently
    0.33
    sometimes
    0.32
     Occasionally
    0.32
    Sometimes
    0.29
    Activations Density 0.358%

    No Known Activations