Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 1
    Prev
    Next
    INDEX
    Explanations

    references to imprisonment and the conditions associated with captivity

    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
    ách
    -0.16
     arrest
    -0.16
    úi
    -0.16
     escort
    -0.16
     arrests
    -0.16
    à¤¾à¤ł
    -0.15
     ÐĿаÑģеление
    -0.15
    iskey
    -0.14
    readcr
    -0.14
    áÄį
    -0.14
    POSITIVE LOGITS
     managed
    0.20
     escape
    0.19
     escaped
    0.19
     Managed
    0.18
     Escape
    0.18
     appeal
    0.17
    escape
    0.17
    Escape
    0.17
     escapes
    0.17
    Managed
    0.16
    Activations Density 0.028%

    No Known Activations