© 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. 12756
    Prev
    Next
    INDEX
    Explanations

    occurrences of the name "Yorick" and references to the fictional character "Lorax."

    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
    mn
    -0.19
    eno
    -0.19
    es
    -0.18
    esian
    -0.18
    esy
    -0.17
    ary
    -0.17
    et
    -0.17
    m
    -0.17
    eder
    -0.16
    enko
    -0.16
    POSITIVE LOGITS
    ror
    0.18
    ourke
    0.18
    rible
    0.18
    ãĥ¼ãĥijãĥ¼
    0.16
    IENTATION
    0.16
    iginal
    0.16
    IENT
    0.16
    leans
    0.15
    outines
    0.15
    ONTAL
    0.15
    Activations Density 0.032%

    No Known Activations