© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 63576
    Prev
    Next
    INDEX
    Explanations

    technical terms and concepts related to API configuration and coding practices

    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
     Fancy
    -0.14
     Rap
    -0.14
    شت
    -0.13
    @testable
    -0.13
     Hans
    -0.13
     подÑģ
    -0.13
    ÙĦÙĬÙĩ
    -0.12
    MethodImpl
    -0.12
    ibrary
    -0.12
    Han
    -0.12
    POSITIVE LOGITS
     configuration
    0.32
     conf
    0.30
     config
    0.29
     JSON
    0.28
     configFile
    0.26
     Configuration
    0.26
    .json
    0.26
     ini
    0.25
     XML
    0.25
     json
    0.25
    Activations Density 0.021%

    No Known Activations