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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 54462
    Prev
    Next
    INDEX
    Explanations

    directives or calls to action

    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
    adow
    -0.16
    dür
    -0.14
    rador
    -0.14
    ayed
    -0.14
    ropolis
    -0.14
    Ĥ¬
    -0.14
    atee
    -0.13
    322
    -0.13
    ylon
    -0.13
    prend
    -0.13
    POSITIVE LOGITS
     yourself
    0.36
     yourselves
    0.29
     Yourself
    0.26
     me
    0.25
     your
    0.24
    ä½łçļĦ
    0.21
    åIJ§
    0.21
    able
    0.20
    ings
    0.19
    ÂŃing
    0.18
    Activations Density 0.309%

    No Known Activations