© 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. 41103
    Prev
    Next
    INDEX
    Explanations

    phrases related to subscribing to newsletters or updates

    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
    saida
    -0.15
    ants
    -0.14
    á»±a
    -0.13
    -0.13
    ftware
    -0.13
    585
    -0.13
    326
    -0.13
     tempered
    -0.12
     ~
    -0.12
    ä½į
    -0.12
    POSITIVE LOGITS
     newsletters
    0.20
    onda
    0.18
     breaking
    0.16
     Breaking
    0.16
     VERIFY
    0.15
     newsletter
    0.15
     subscriber
    0.15
     Patch
    0.14
    esson
    0.14
    taboola
    0.14
    Activations Density 2.892%

    No Known Activations