© 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. 44478
    Prev
    Next
    INDEX
    Explanations

    phrases related to plans and ambitions

    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
    IODevice
    -0.14
    adu
    -0.14
    avad
    -0.14
    ζί
    -0.14
    .Override
    -0.14
    clud
    -0.13
    ýn
    -0.13
    .tbl
    -0.13
     kdyby
    -0.13
    Ñħови
    -0.13
    POSITIVE LOGITS
     accompl
    0.65
     accomplish
    0.64
     accomplished
    0.53
     achieve
    0.47
     accomplishment
    0.45
     achieves
    0.43
     Achie
    0.42
     achieving
    0.40
    Achie
    0.39
     achie
    0.39
    Activations Density 0.044%

    No Known Activations