© 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. 63528
    Prev
    Next
    INDEX
    Explanations

    references to past actions or states of being

    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
     removeFromSuperview
    -0.16
    OrFail
    -0.16
    iful
    -0.14
     Kb
    -0.14
     finally
    -0.14
    inker
    -0.14
    _REGISTRY
    -0.13
    ãĥ¼ãĥī
    -0.13
    ahun
    -0.13
     Loose
    -0.13
    POSITIVE LOGITS
     suspension
    0.67
     suspend
    0.64
     suspended
    0.61
     susp
    0.61
     Suspension
    0.60
     Susp
    0.52
     pause
    0.50
    Susp
    0.48
    suspend
    0.47
     paused
    0.45
    Activations Density 0.045%

    No Known Activations