© 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. 54192
    Prev
    Next
    INDEX
    Explanations

    phrases related to emotional or psychological states and personal experiences

    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
    aha
    -0.16
    isia
    -0.15
    áº
    -0.15
    ambre
    -0.14
    amina
    -0.14
    ãģłãģ£ãģ¦
    -0.14
    ener
    -0.14
    etter
    -0.14
    atus
    -0.13
    uma
    -0.13
    POSITIVE LOGITS
     slightly
    0.73
     somewhat
    0.72
     slight
    0.43
     немного
    0.41
     biraz
    0.40
     bit
    0.39
     trochu
    0.37
     ÑĤÑĢоÑħи
    0.36
     chút
    0.36
     slightest
    0.34
    Activations Density 0.104%

    No Known Activations