© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 63642
    Prev
    Next
    INDEX
    Explanations

    mentions of users and complaints in various contexts

    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
    èŃĺ
    -0.17
    urette
    -0.15
    VML
    -0.14
    oram
    -0.14
    acock
    -0.14
    _Abstract
    -0.14
    emek
    -0.14
    phies
    -0.14
    .Undef
    -0.14
    aber
    -0.14
    POSITIVE LOGITS
     report
    0.62
     reporting
    0.61
     reported
    0.61
     reports
    0.55
     Reporting
    0.49
    report
    0.49
    reported
    0.47
     Report
    0.47
    Report
    0.46
    -reported
    0.45
    Activations Density 0.036%

    No Known Activations