© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 41020
    Prev
    Next
    INDEX
    Explanations

    instances of the word "added" along with supporting phrases that convey opinion or commentary

    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
     addition
    -0.18
     additions
    -0.17
     Addition
    -0.16
     additives
    -0.16
    _Api
    -0.15
    _add
    -0.15
    adding
    -0.15
    loo
    -0.15
    .opend
    -0.14
     added
    -0.14
    POSITIVE LOGITS
    uce
    0.20
    endum
    0.19
    swer
    0.18
    iten
    0.16
     insult
    0.16
    ictions
    0.15
    ConverterFactory
    0.15
     later
    0.15
    erase
    0.15
    iction
    0.15
    Activations Density 0.016%

    No Known Activations