© 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. 15076
    Prev
    Next
    INDEX
    Explanations

    instances of judgment and evaluation regarding people or situations

    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
    ONT
    -0.15
    ednou
    -0.15
    diet
    -0.14
    idel
    -0.14
     belts
    -0.13
    angan
    -0.13
    веÑī
    -0.13
    ónico
    -0.13
    phin
    -0.13
     claro
    -0.13
    POSITIVE LOGITS
     based
    0.67
    based
    0.55
     Based
    0.53
    Based
    0.47
     according
    0.46
     dá»±a
    0.41
    _based
    0.41
    according
    0.37
    åŁºäºİ
    0.34
     According
    0.33
    Activations Density 0.046%

    No Known Activations