© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 33080
    Prev
    Next
    INDEX
    Explanations

    phrases indicating shared attributes or commonalities among groups or ideas

    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
    Monad
    -0.15
    lop
    -0.15
    ycl
    -0.15
    åħ¥ãĤĮ
    -0.14
    ÑĸÑģно
    -0.14
    inci
    -0.14
     manner
    -0.14
    ovel
    -0.13
    206
    -0.13
    lys
    -0.13
    POSITIVE LOGITS
     denominator
    0.38
     denomin
    0.33
     thread
    0.32
    ality
    0.30
    /shared
    0.30
    -den
    0.27
     shared
    0.26
     THREAD
    0.26
     Den
    0.25
    thread
    0.25
    Activations Density 0.015%

    No Known Activations