© 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. 29730
    Prev
    Next
    INDEX
    Explanations

    occurrences of indicators of location and involvement in actions

    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
     both
    -0.30
    both
    -0.28
     Both
    -0.26
    Both
    -0.26
     everywhere
    -0.25
     både
    -0.23
     BOTH
    -0.22
     everything
    -0.21
     beide
    -0.21
    ãģĻãģ¹ãģ¦
    -0.20
    POSITIVE LOGITS
     either
    0.86
     EITHER
    0.69
    either
    0.68
     any
    0.67
     Either
    0.66
     ANY
    0.63
     anything
    0.61
     anyone
    0.60
    Either
    0.59
     anybody
    0.57
    Activations Density 0.360%

    No Known Activations