© 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. 15620
    Prev
    Next
    INDEX
    Explanations

    terms related to expression or acts of expressing oneself

    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
    åĪĴ
    -0.16
    ffects
    -0.16
    quet
    -0.15
    ahren
    -0.15
    ffect
    -0.14
    <message
    -0.14
    uteur
    -0.14
    sters
    -0.14
    ม
    -0.14
    Ñħи
    -0.14
    POSITIVE LOGITS
    ively
    0.26
    iveness
    0.23
    ions
    0.20
    ible
    0.19
    à¸ģารà¸ĵ
    0.18
    ional
    0.18
    ive
    0.18
    son
    0.17
    ivism
    0.17
    (Expression
    0.16
    Activations Density 0.020%

    No Known Activations