© 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. 24387
    Prev
    Next
    INDEX
    Explanations

    text related to educational or disciplinary actions taken on students or teachers

    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
    ÑĤÑĢо
    -0.17
    svp
    -0.16
    IOD
    -0.15
    ikip
    -0.15
    onga
    -0.15
    ponsored
    -0.15
    AAD
    -0.14
    ibold
    -0.14
    exion
    -0.14
    ikit
    -0.14
    POSITIVE LOGITS
     teacher
    0.85
     teachers
    0.84
     Teachers
    0.76
     Teacher
    0.75
     teaching
    0.67
    teacher
    0.65
    teachers
    0.65
    Teacher
    0.64
    Teachers
    0.64
     Teaching
    0.57
    Activations Density 0.032%

    No Known Activations