© 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. 44127
    Prev
    Next
    INDEX
    Explanations

    numerical data and color 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
    ác
    -0.14
    uen
    -0.14
    621
    -0.14
    owing
    -0.13
    legg
    -0.13
    пи
    -0.13
    plib
    -0.13
    edException
    -0.12
    lice
    -0.12
     Of
    -0.12
    POSITIVE LOGITS
    -and
    0.91
    _and
    0.56
    And
    0.49
    -&
    0.46
    _AND
    0.45
     And
    0.45
     &
    0.41
    &R
    0.39
    &
    0.39
    &D
    0.38
    Activations Density 0.039%

    No Known Activations