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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 15572
    Prev
    Next
    INDEX
    Explanations

    event details related to performances or shows

    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
    YG
    -0.18
    ailer
    -0.16
    ãĥ³ãĥģ
    -0.16
    odesk
    -0.15
    ALA
    -0.15
     gol
    -0.15
    enheim
    -0.14
    achat
    -0.14
    ynes
    -0.14
    ogle
    -0.14
    POSITIVE LOGITS
     cab
    0.29
     Impro
    0.29
     improv
    0.28
     spoken
    0.26
    Impro
    0.24
     Cab
    0.24
     bur
    0.23
     stand
    0.23
     acts
    0.22
     mime
    0.21
    Activations Density 0.033%

    No Known Activations