© 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. 13188
    Prev
    Next
    INDEX
    Explanations

    phrases indicating desires, intentions, or future actions

    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
    alls
    -0.16
    udiant
    -0.15
    VAS
    -0.15
    icios
    -0.14
     understanding
    -0.14
    anton
    -0.14
     only
    -0.14
    loh
    -0.14
    cip
    -0.14
    iner
    -0.14
    POSITIVE LOGITS
     accomplish
    0.32
     accomplished
    0.31
     happen
    0.28
     happening
    0.27
     accompl
    0.27
     happens
    0.25
     happened
    0.24
     aconte
    0.24
     accomplishment
    0.24
     achieve
    0.23
    Activations Density 0.036%

    No Known Activations