© 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. 43663
    Prev
    Next
    INDEX
    Explanations

    phrases indicating significant performance issues or failures

    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
     metav
    -0.15
    rend
    -0.15
    995
    -0.14
    اصر
    -0.14
    ahat
    -0.14
    ç¬
    -0.13
    empo
    -0.13
     Ðļоли
    -0.13
    æĵ¦
    -0.13
    lue
    -0.13
    POSITIVE LOGITS
     mess
    0.30
     disaster
    0.28
     Disaster
    0.22
     joke
    0.21
    mess
    0.20
     ver
    0.20
     mixed
    0.19
     jokes
    0.18
     disappointment
    0.18
     train
    0.18
    Activations Density 0.073%

    No Known Activations