© 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. 63617
    Prev
    Next
    INDEX
    Explanations

    conditional phrases and concerns about 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
     finally
    -0.15
     Retro
    -0.14
    oreach
    -0.14
     finalize
    -0.14
     Finally
    -0.13
     slowly
    -0.13
    finally
    -0.13
     ultimately
    -0.13
    olucion
    -0.13
    izacion
    -0.13
    POSITIVE LOGITS
     again
    0.69
    again
    0.56
     Again
    0.52
    Again
    0.51
     repeat
    0.51
     another
    0.49
    åĨį
    0.48
     wieder
    0.45
     repeating
    0.43
     novamente
    0.43
    Activations Density 0.053%

    No Known Activations