© 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. 15142
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "ere" and its variations, as well as context related to financial or legal situations

    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
    re
    -0.20
    ri
    -0.18
    resp
    -0.17
    da
    -0.17
    rie
    -0.16
    ra
    -0.16
    sko
    -0.16
    stadt
    -0.15
    reat
    -0.15
    mand
    -0.15
    POSITIVE LOGITS
    ignty
    0.18
    ults
    0.17
    anou
    0.17
    ira
    0.16
    gow
    0.15
    zo
    0.15
    Ïħν
    0.15
    Ïħνα
    0.15
    olas
    0.15
    eneg
    0.15
    Activations Density 0.007%

    No Known Activations