© 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. 21619
    Prev
    Next
    INDEX
    Explanations

    mentions of "results" or similar terms related to outcomes and their significance

    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
     resulting
    -0.19
    Result
    -0.19
    thing
    -0.18
     resultant
    -0.17
    _result
    -0.17
     resulted
    -0.17
    result
    -0.16
    oria
    -0.15
     result
    -0.15
    roll
    -0.15
    POSITIVE LOGITS
    ados
    0.24
    antly
    0.22
    물
    0.19
    ative
    0.19
    물ìĿĦ
    0.18
    -oriented
    0.18
    ants
    0.18
    DDS
    0.17
    oriented
    0.16
     obtained
    0.16
    Activations Density 0.035%

    No Known Activations