© 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. 25157
    Prev
    Next
    INDEX
    Explanations

    mentions of actions taken by a company related to technology and product development

    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
    sville
    -0.17
    CEF
    -0.17
    oldem
    -0.16
    eger
    -0.16
    loff
    -0.14
    cea
    -0.14
    __["
    -0.14
    vet
    -0.14
    utherford
    -0.13
    jas
    -0.13
    POSITIVE LOGITS
     set
    0.87
     sets
    0.61
    set
    0.59
     Set
    0.57
    (set
    0.55
    -set
    0.52
    .set
    0.51
    _set
    0.49
    =set
    0.48
    ãĤ»ãĥĥãĥĪ
    0.48
    Activations Density 0.028%

    No Known Activations