© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 44934
    Prev
    Next
    INDEX
    Explanations

    instances of deception, manipulation, and ambiguity in language

    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
    ayar
    -0.14
     táºŃn
    -0.13
    üny
    -0.12
    indered
    -0.12
    dge
    -0.12
    .myapplication
    -0.12
    ̣
    -0.12
     Infer
    -0.12
    ificantly
    -0.12
    enz
    -0.12
    POSITIVE LOGITS
    instead
    0.19
     instead
    0.15
     вмеÑģÑĤ
    0.15
     Instead
    0.15
    Instead
    0.14
    fragistics
    0.14
    åī²
    0.14
    ÑīинÑĭ
    0.14
    htdocs
    0.14
    isms
    0.14
    Activations Density 0.160%

    No Known Activations