© 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. 23055
    Prev
    Next
    INDEX
    Explanations

    instances of the word "nearly" in relation to various contexts

    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
    ifton
    -0.18
    earing
    -0.17
    丶
    -0.14
    eted
    -0.14
     nghi
    -0.14
    regunta
    -0.14
    imer
    -0.14
    iform
    -0.14
    ØŃاد
    -0.14
    ferred
    -0.14
    POSITIVE LOGITS
     always
    0.19
    ร
    0.19
     everyone
    0.18
     dozen
    0.18
     impossible
    0.18
    100
    0.17
     identical
    0.17
     all
    0.17
    ness
    0.17
     every
    0.17
    Activations Density 0.030%

    No Known Activations