© 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. 54498
    Prev
    Next
    INDEX
    Explanations

    references to physical size or characteristics, particularly hands

    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
    engers
    -0.17
     Ders
    -0.16
    UIBarButtonItem
    -0.16
    erten
    -0.15
    çĹħ
    -0.15
    ildo
    -0.15
    stamp
    -0.15
    åĨĴ
    -0.14
    arov
    -0.14
    _VC
    -0.14
    POSITIVE LOGITS
     fighters
    0.32
     ring
    0.31
     boxing
    0.31
     training
    0.29
     punches
    0.28
     punching
    0.28
     fighter
    0.28
     fight
    0.28
    -ring
    0.27
     rings
    0.27
    Activations Density 0.024%

    No Known Activations