© 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. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 8515
    Prev
    Next
    INDEX
    Explanations

    keywords related to drug overdoses and addiction treatment

    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
    ãĥ¼ãĥ³
    -0.15
    ÌĢ
    -0.15
    łģ
    -0.14
    ìĨ
    -0.14
    riba
    -0.14
     embarrassing
    -0.14
     Antar
    -0.14
    emm
    -0.14
    znik
    -0.14
     Mond
    -0.14
    POSITIVE LOGITS
     synthetic
    0.29
     Synthetic
    0.25
     powerful
    0.22
     synth
    0.21
     potent
    0.21
    syn
    0.21
    Syn
    0.19
     strong
    0.19
     Powerful
    0.18
     synthesized
    0.18
    Activations Density 0.024%

    No Known Activations