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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 42248
    Prev
    Next
    INDEX
    Explanations

    the adjective "big" and its related contexts in sentences

    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
    opsis
    -0.15
    itore
    -0.15
    วà¸ĩ
    -0.14
    ç²¾åĵģ
    -0.14
    /stdc
    -0.14
    amentals
    -0.14
    nett
    -0.14
    ÅĤem
    -0.14
    ÑĢÑĸб
    -0.14
    esis
    -0.14
    POSITIVE LOGITS
    -time
    0.20
    time
    0.20
    -name
    0.18
    gest
    0.17
    fect
    0.16
    GEST
    0.16
    ertz
    0.16
    gie
    0.16
    oted
    0.15
    å½¹
    0.15
    Activations Density 0.022%

    No Known Activations