© 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. 22113
    Prev
    Next
    INDEX
    Explanations

    references to safety systems and equipment related to public transportation

    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
    amac
    -0.16
    enis
    -0.15
    .scalablytyped
    -0.15
    .fhir
    -0.14
    weetalert
    -0.14
    ParticleSystem
    -0.14
    надлеж
    -0.14
    PUR
    -0.13
    .SizeF
    -0.13
    Pending
    -0.13
    POSITIVE LOGITS
     Pro
    0.97
     pro
    0.85
    Pro
    0.80
    pro
    0.74
    _pro
    0.69
     PRO
    0.65
    -Pro
    0.64
    .pro
    0.64
     proton
    0.63
    (pro
    0.63
    Activations Density 0.012%

    No Known Activations