© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 30554
    Prev
    Next
    INDEX
    Explanations

    phrases indicating potential actions or outcomes

    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
    ':''
    -0.16
    aits
    -0.15
    elves
    -0.14
     evet
    -0.14
    ayıp
    -0.14
    ADDING
    -0.14
     gó
    -0.14
    _Utils
    -0.13
    RTOS
    -0.13
    lore
    -0.13
    POSITIVE LOGITS
     '
    0.28
     ‘
    0.22
     fu
    0.19
     "
    0.18
     risk
    0.18
     finally
    0.17
     linked
    0.17
     shield
    0.17
     face
    0.17
     soon
    0.17
    Activations Density 0.133%

    No Known Activations