© 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. 30097
    Prev
    Next
    INDEX
    Explanations

    terms associated with dehumanization and its implications

    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
    ALLY
    -0.17
    ì§
    -0.16
    ì§Ħ
    -0.16
    çľī
    -0.15
    inya
    -0.15
    urally
    -0.15
    ,eg
    -0.15
    adiens
    -0.14
    teki
    -0.14
     Trie
    -0.14
    POSITIVE LOGITS
    ization
    0.21
    ification
    0.18
    ibration
    0.18
    .biz
    0.18
    íĻĶ
    0.16
    auc
    0.16
    ing
    0.16
    ermen
    0.15
    tlement
    0.15
    ypse
    0.15
    Activations Density 0.055%

    No Known Activations