© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 29283
    Prev
    Next
    INDEX
    Explanations

    the word "unless."

    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
    agos
    -0.16
    OME
    -0.14
    rani
    -0.14
    jo
    -0.14
    ãģ©
    -0.13
    udic
    -0.13
     neither
    -0.13
    ledged
    -0.13
    erken
    -0.13
    ansa
    -0.13
    POSITIVE LOGITS
    /un
    0.37
     otherwise
    0.32
     OTHERWISE
    0.24
    otherwise
    0.22
     Otherwise
    0.22
     there
    0.20
     someone
    0.20
    Otherwise
    0.20
     unless
    0.19
     they
    0.18
    Activations Density 0.036%

    No Known Activations