© 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. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 21439
    Prev
    Next
    INDEX
    Explanations

    prepositions and terms related to instructions and conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     someone
    -0.24
     somebody
    -0.21
    someone
    -0.20
     something
    -0.17
     or
    -0.16
     Someone
    -0.16
     another
    -0.16
     elsewhere
    -0.16
    Others
    -0.16
    something
    -0.15
    POSITIVE LOGITS
     all
    0.69
    æīĢæľī
    0.54
     semua
    0.49
    ãģĻãģ¹ãģ¦
    0.48
     wszyst
    0.47
     ALL
    0.45
     вÑģеÑħ
    0.45
     모ëĵł
    0.43
     tất
    0.43
     ÏĮλ
    0.41
    Activations Density 0.094%

    No Known Activations