© 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. 44164
    Prev
    Next
    INDEX
    Explanations

    phrases related to unfamiliar or uncharted experiences and territories

    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
    ÑĪки
    -0.14
    ิà¸ĩ
    -0.14
    idal
    -0.14
    upal
    -0.14
    åĬĽçļĦ
    -0.14
    rored
    -0.13
    apel
    -0.13
    letcher
    -0.13
     Toolbox
    -0.13
     Continental
    -0.13
    POSITIVE LOGITS
     territory
    0.98
     Territory
    0.78
     territories
    0.74
     terrain
    0.71
    territ
    0.71
     territor
    0.62
     terr
    0.59
    Terr
    0.58
     ground
    0.57
     Territories
    0.57
    Activations Density 0.029%

    No Known Activations