© 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. 21663
    Prev
    Next
    INDEX
    Explanations

    phrases indicating source or origin

    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
    ãģijãĤĮãģ©
    -0.15
    опаÑģ
    -0.14
    _firestore
    -0.14
    .Sdk
    -0.14
    ราย
    -0.14
    кап
    -0.14
    лÑıÑħ
    -0.14
    \Carbon
    -0.14
    .ObjectId
    -0.14
    ä»İ
    -0.13
    POSITIVE LOGITS
    /to
    0.38
     across
    0.31
    /about
    0.29
     within
    0.28
     scratch
    0.25
     around
    0.25
    t
    0.25
     sources
    0.24
     abroad
    0.23
     inside
    0.23
    Activations Density 0.236%

    No Known Activations