© 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. 41952
    Prev
    Next
    INDEX
    Explanations

    negations or phrases that express a denial or contradiction

    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
     nackte
    -0.17
    thon
    -0.16
     OTHERWISE
    -0.16
     noen
    -0.16
    itore
    -0.14
    #
    -0.14
    ELSE
    -0.14
    ndo
    -0.14
    ĤŃ
    -0.13
     treff
    -0.13
    POSITIVE LOGITS
     yet
    0.26
     unless
    0.24
    yet
    0.20
    Yet
    0.19
     anymore
    0.19
    withstanding
    0.18
    ional
    0.18
     until
    0.18
     Yet
    0.18
    unless
    0.17
    Activations Density 0.024%

    No Known Activations