© 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 EvalsExportsAPI 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. 43023
    Prev
    Next
    INDEX
    Explanations

    phrases related to personal opinions and changes in a system

    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
    imest
    -0.15
    utz
    -0.15
     gonna
    -0.15
    —are
    -0.15
    won
    -0.14
    assic
    -0.14
    may
    -0.14
    hots
    -0.14
    —is
    -0.14
    errMsg
    -0.13
    POSITIVE LOGITS
     would
    0.55
    would
    0.42
     Would
    0.41
    Would
    0.38
     zou
    0.34
     würde
    0.29
    OULD
    0.28
    'd
    0.27
     serait
    0.27
     skulle
    0.26
    Activations Density 0.098%

    No Known Activations