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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 33324
    Prev
    Next
    INDEX
    Explanations

    phrases describing physical and tangible objects or experiences

    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
    toi
    -0.14
    IBUTE
    -0.14
     compact
    -0.14
     à¤ļà¤ķ
    -0.14
    बर
    -0.14
    ehler
    -0.14
     Pace
    -0.14
    èĵ
    -0.14
    ìĿ´íģ¬
    -0.13
    739
    -0.13
    POSITIVE LOGITS
     hang
    0.85
     hanging
    0.81
     hangs
    0.80
     Hang
    0.76
     hung
    0.75
     Hanging
    0.73
    Hang
    0.72
    hang
    0.71
     suspended
    0.54
     dangling
    0.51
    Activations Density 0.038%

    No Known Activations