© 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. 28838
    Prev
    Next
    INDEX
    Explanations

    references to items and components related to craftsmanship and product quality

    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
     of
    -0.17
    lessly
    -0.14
    oe
    -0.14
     taj
    -0.14
    [d
    -0.13
     option
    -0.13
     into
    -0.13
    option
    -0.13
    preci
    -0.13
    idth
    -0.13
    POSITIVE LOGITS
     itself
    0.35
    's
    0.23
    ’s
    0.22
    åıĬåħ¶
    0.21
    osphere
    0.17
    ův
    0.16
    .AppendFormat
    0.16
    .***.***
    0.16
    ’S
    0.16
    WARDED
    0.15
    Activations Density 0.479%

    No Known Activations