© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 16-LLAMASCOPE-RES-32K
    11. 7099
    Prev
    Next
    INDEX
    Explanations

    terms related to medical care and recovery from injuries, particularly concussions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     crises
    -0.16
    olith
    -0.16
     rouge
    -0.15
     Sole
    -0.15
     sick
    -0.15
    ãĥ¬ãĥĥãĥĪ
    -0.15
     sinking
    -0.14
    izza
    -0.14
    iken
    -0.14
     cruc
    -0.13
    POSITIVE LOGITS
     permanently
    0.31
     permanent
    0.30
    æ°¸ä¹ħ
    0.29
     forever
    0.26
     Permanent
    0.26
    Permanent
    0.25
    permanent
    0.25
     perman
    0.23
     Forever
    0.23
     scar
    0.23
    Activations Density 0.196%

    No Known Activations