Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 23215
    Prev
    Next
    INDEX
    Explanations

    mentions of the letter "L" as it appears in 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
    ink
    -0.19
    ane
    -0.18
    inea
    -0.18
    in
    -0.16
    ine
    -0.16
    PCS
    -0.16
    ines
    -0.16
    aw
    -0.16
     ups
    -0.15
    ÙĤ
    -0.15
    POSITIVE LOGITS
    alu
    0.22
    etic
    0.20
    el
    0.19
    ette
    0.18
    ulu
    0.18
    ucc
    0.17
    oka
    0.17
    iveness
    0.17
    oris
    0.16
    op
    0.16
    Activations Density 0.048%

    No Known Activations