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. 23581
    Prev
    Next
    INDEX
    Explanations

    numerical values and percentages

    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
    aż
    -0.14
    atu
    -0.14
    pad
    -0.13
     bec
    -0.13
    uze
    -0.13
    etty
    -0.13
     ÙħØ·
    -0.12
    umper
    -0.12
    eny
    -0.12
    å½¹
    -0.12
    POSITIVE LOGITS
    æĸĻçĦ¡æĸĻ
    0.15
    éĺħ读次æķ°
    0.15
    OffsetTable
    0.14
    езÑĥлÑĮÑĤ
    0.14
    _Tis
    0.14
    _tF
    0.14
    _mC
    0.14
    UsageId
    0.14
    _mD
    0.14
    .bundle
    0.13
    Activations Density 0.025%

    No Known Activations