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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 13276
    Prev
    Next
    INDEX
    Explanations

    references to wildlife and environmental concerns

    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
    çķ
    -0.15
    лÑıд
    -0.14
    ertz
    -0.14
     sheep
    -0.14
    oya
    -0.14
    ewise
    -0.14
     speculative
    -0.13
    γνÏī
    -0.13
    *)((
    -0.13
     Shields
    -0.13
    POSITIVE LOGITS
     trap
    0.24
     radio
    0.22
     PIT
    0.22
     Trap
    0.21
     traps
    0.21
     citizen
    0.20
     camera
    0.20
     collar
    0.19
    Trap
    0.19
     rele
    0.19
    Activations Density 0.018%

    No Known Activations