© 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. 63197
    Prev
    Next
    INDEX
    Explanations

    verbs associated with actions or processes related to giving, sending, or transferring items or information

    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
    AV
    -0.14
     réfé
    -0.14
    _DLL
    -0.14
    anda
    -0.14
    akis
    -0.14
    stay
    -0.13
    /from
    -0.13
    ãĤ»ãĥ³
    -0.13
    ators
    -0.13
    scratch
    -0.13
    POSITIVE LOGITS
     it
    0.42
     them
    0.34
    å®ĥ
    0.23
    them
    0.23
     ìĿ´ë¥¼
    0.22
    å®ĥ们
    0.22
    ãģĿãĤĮãģ¯
    0.22
     nó
    0.21
    ãģĿãĤĮ
    0.21
     chúng
    0.20
    Activations Density 0.089%

    No Known Activations