© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 25418
    Prev
    Next
    INDEX
    Explanations

    occurrences of numeric and punctuation symbols indicating social media references

    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
     Summers
    -0.18
     nat
    -0.18
     Easter
    -0.18
     summers
    -0.17
     Summer
    -0.17
    nat
    -0.17
     summer
    -0.16
     Natal
    -0.16
    Summer
    -0.16
     Natasha
    -0.16
    POSITIVE LOGITS
     oct
    0.87
     Oct
    0.86
    Oct
    0.80
     October
    0.80
    October
    0.76
    _oct
    0.73
    oct
    0.72
     OCT
    0.70
    /oct
    0.69
    -Oct
    0.66
    Activations Density 0.005%

    No Known Activations