© 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. 42378
    Prev
    Next
    INDEX
    Explanations

    discourse surrounding expectations and responsibilities in professional or academic contexts

    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
    ighton
    -0.15
     Dag
    -0.14
    cpt
    -0.14
    ADDE
    -0.14
    Printf
    -0.13
    apis
    -0.13
     Dawson
    -0.13
    dal
    -0.13
     Dak
    -0.13
    emachine
    -0.13
    POSITIVE LOGITS
     doing
    0.90
     Doing
    0.79
    doing
    0.79
    Doing
    0.74
    åģļ
    0.68
     do
    0.63
     done
    0.58
     does
    0.54
     did
    0.52
    à¸Ĺำ
    0.46
    Activations Density 0.213%

    No Known Activations