© 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. 13180
    Prev
    Next
    INDEX
    Explanations

    phrases related to project development and management

    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
    coni
    -0.15
     Milton
    -0.14
    itti
    -0.14
    ovice
    -0.14
    StringEncoding
    -0.13
    าสà¸ķร
    -0.13
    mw
    -0.12
    ensa
    -0.12
    antee
    -0.12
     Lama
    -0.12
    POSITIVE LOGITS
     management
    1.06
     manage
    1.02
     Management
    0.96
     managing
    0.91
     manager
    0.91
    管çIJĨ
    0.90
     Manage
    0.88
     manages
    0.88
     managers
    0.87
    -management
    0.84
    Activations Density 0.007%

    No Known Activations