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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 12501
    Prev
    Next
    INDEX
    Explanations

    phrases related to character evaluation and strategic recommendations

    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
    OrFail
    -0.16
    ä»ĭ
    -0.15
    Chip
    -0.15
    ruba
    -0.15
     involvement
    -0.15
     Chip
    -0.15
    isure
    -0.14
    fx
    -0.14
    ÑĢин
    -0.14
    alten
    -0.14
    POSITIVE LOGITS
     play
    0.20
     max
    0.19
     plays
    0.18
    play
    0.18
    ocket
    0.17
     equip
    0.17
     slee
    0.17
    plays
    0.17
     farming
    0.17
    403
    0.17
    Activations Density 0.052%

    No Known Activations