© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 29170
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing ownership or possession

    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
    ãģıãĤĭ
    -0.20
     fell
    -0.20
     underwent
    -0.18
    _was
    -0.18
     Came
    -0.18
     grew
    -0.17
     Was
    -0.17
    came
    -0.17
    fell
    -0.16
     arose
    -0.16
    POSITIVE LOGITS
     been
    0.46
     gone
    0.44
     gotten
    0.38
     Been
    0.38
    been
    0.36
    gone
    0.35
     BEEN
    0.34
    Been
    0.32
     Gone
    0.32
     given
    0.30
    Activations Density 0.074%

    No Known Activations