© 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. 24719
    Prev
    Next
    INDEX
    Explanations

    references to phone conversations and related interactions

    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
     over
    -0.17
    á»įt
    -0.15
    elter
    -0.15
    izr
    -0.15
    isci
    -0.15
    ascus
    -0.14
    256
    -0.14
    alc
    -0.14
    620
    -0.14
    owitz
    -0.14
    POSITIVE LOGITS
     phone
    0.29
    phone
    0.25
    -phone
    0.23
    .phone
    0.22
    Phone
    0.22
     PHONE
    0.21
     line
    0.21
     Phone
    0.21
    éĽ»è©±
    0.19
    çĶµè¯Ŀ
    0.19
    Activations Density 0.013%

    No Known Activations