© 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. 63634
    Prev
    Next
    INDEX
    Explanations

    conjunctions and phrases that express a balance of pros and cons

    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
    .onResume
    -0.15
    ansen
    -0.15
    ượng
    -0.15
    _GRANTED
    -0.14
     âĸ²
    -0.14
    ffect
    -0.14
    íĸ¥
    -0.14
    rane
    -0.14
    andra
    -0.13
     vict
    -0.13
    POSITIVE LOGITS
     convers
    0.31
     naopak
    0.27
     vice
    0.26
     Conversely
    0.25
    NegativeButton
    0.24
     negative
    0.23
     negatives
    0.22
     detr
    0.22
     drawback
    0.21
     downside
    0.20
    Activations Density 0.069%

    No Known Activations