© 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. 13407
    Prev
    Next
    INDEX
    Explanations

    instances of the word "straightforward" and its variations

    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
    sein
    -0.17
    hop
    -0.17
     streets
    -0.16
     street
    -0.16
    sey
    -0.16
     storm
    -0.16
    istrict
    -0.15
     standards
    -0.15
    Äįem
    -0.15
    ymes
    -0.15
    POSITIVE LOGITS
    ened
    0.23
    ening
    0.21
    ÙĪØ·
    0.18
    -ÑĤаки
    0.18
    pling
    0.18
    ener
    0.17
    visor
    0.17
    heets
    0.17
    pile
    0.17
    eners
    0.17
    Activations Density 0.016%

    No Known Activations