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. 23575
    Prev
    Next
    INDEX
    Explanations

    references to specific locations and contextual details related to events and regulations

    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
    â̦but
    -0.29
    â̦and
    -0.29
    â̦the
    -0.24
    â̦I
    -0.22
    —but
    -0.22
    â̦it
    -0.22
    —I
    -0.21
    —if
    -0.20
    —they
    -0.20
    —and
    -0.20
    POSITIVE LOGITS
    ©©
    0.21
    çĽijåIJ¬é¡µéĿ¢
    0.18
    .LayoutStyle
    0.16
    çͳåįļ
    0.15
    ä¸ĭ载次æķ°
    0.15
    ç½ijåĪĬ
    0.15
    太éĺ³åŁİ
    0.14
    .datatables
    0.14
    Äł
    0.14
    .yy
    0.14
    Activations Density 0.163%

    No Known Activations