© 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. 41806
    Prev
    Next
    INDEX
    Explanations

    phrases and notions related to accountability and responsibilities

    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
     obl
    -0.15
     reminis
    -0.15
     ακÏĮ
    -0.15
     ëĤĺê°Ģ
    -0.15
    lest
    -0.14
     troub
    -0.14
     encounter
    -0.14
    isis
    -0.14
     wee
    -0.14
    hit
    -0.14
    POSITIVE LOGITS
     being
    0.28
     having
    0.25
     mir
    0.23
     making
    0.22
     associ
    0.22
     bas
    0.21
     isol
    0.21
     taking
    0.21
     correl
    0.20
     attrib
    0.19
    Activations Density 0.526%

    No Known Activations