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. 68
    Prev
    Next
    INDEX
    Explanations

    references to specific filenames or identifiers in code or data 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
     Ashe
    -0.15
    伯
    -0.15
     Altın
    -0.14
    iyel
    -0.14
    .EventType
    -0.14
     LENG
    -0.14
    ÑĨо
    -0.14
     hôm
    -0.14
    iyon
    -0.14
    xEB
    -0.14
    POSITIVE LOGITS
    ita
    0.23
    ela
    0.22
    enda
    0.21
    ella
    0.21
    unda
    0.20
    ia
    0.20
    ina
    0.20
    ta
    0.20
    ota
    0.20
    ya
    0.19
    Activations Density 0.021%

    No Known Activations