© 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. 63262
    Prev
    Next
    INDEX
    Explanations

    the core themes and emotional depth within narratives

    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
    åĽłæŃ¤
    -0.17
     therefore
    -0.16
     consequently
    -0.15
    -LAST
    -0.15
     allegedly
    -0.15
     thus
    -0.15
     thereby
    -0.15
    	then
    -0.14
    Therefore
    -0.14
    DBus
    -0.14
    POSITIVE LOGITS
     definitely
    0.34
     actually
    0.28
     Definitely
    0.25
     certainly
    0.24
    actually
    0.22
    !
    0.21
    laughs
    0.20
     indeed
    0.20
     Actually
    0.19
     yes
    0.19
    Activations Density 0.754%

    No Known Activations