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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 30582
    Prev
    Next
    INDEX
    Explanations

    measurements and scores related to rankings and evaluations, particularly in academic or medical contexts

    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
    ARAM
    -0.15
    enson
    -0.15
     Vance
    -0.15
    QualifiedName
    -0.14
    PageSize
    -0.14
    engkap
    -0.13
     census
    -0.13
    AREST
    -0.13
    ãĥ
    -0.13
    oyal
    -0.13
    POSITIVE LOGITS
     score
    1.00
     scores
    0.87
     Score
    0.83
    score
    0.79
    Score
    0.78
    -score
    0.74
     Scores
    0.73
    _score
    0.72
    .score
    0.68
     SCORE
    0.66
    Activations Density 0.062%

    No Known Activations