© 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. 55417
    Prev
    Next
    INDEX
    Explanations

    references to biological processes and interactions involving genes and immune responses

    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
    综åIJĪ
    -0.14
    æij©
    -0.14
    emoc
    -0.14
    INGER
    -0.14
     Jew
    -0.13
    ioc
    -0.13
    istros
    -0.13
     upgrade
    -0.12
    Ãły
    -0.12
    /logger
    -0.12
    POSITIVE LOGITS
     human
    0.18
     wild
    0.17
     Fox
    0.17
     prohibit
    0.15
    .si
    0.15
    human
    0.15
    /Area
    0.15
     cas
    0.15
    wild
    0.15
     expression
    0.14
    Activations Density 0.047%

    No Known Activations