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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8973
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a return to a previous location or state

    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
    anine
    -0.15
    new
    -0.15
    --+
    -0.14
    è¦ļ
    -0.14
    an
    -0.14
    anut
    -0.13
    ledon
    -0.13
     responseBody
    -0.13
    ütün
    -0.13
    erb
    -0.13
    POSITIVE LOGITS
     normal
    0.33
    normal
    0.25
     whence
    0.24
     sender
    0.23
     home
    0.22
    -normal
    0.22
    æŃ£å¸¸
    0.22
     NORMAL
    0.21
     where
    0.21
     square
    0.21
    Activations Density 0.049%

    No Known Activations