© 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. 21696
    Prev
    Next
    INDEX
    Explanations

    subjective judgments and evaluations of concepts or situations

    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
     Could
    -0.20
    Could
    -0.18
     potential
    -0.18
    potential
    -0.17
     Potential
    -0.17
    avax
    -0.16
    ë¦
    -0.16
    ìĿ¸ëį°
    -0.15
     поÑĤен
    -0.14
    longleftrightarrow
    -0.14
    POSITIVE LOGITS
     may
    0.96
    may
    0.78
     might
    0.66
     MAY
    0.54
    might
    0.52
     можеÑĤ
    0.45
     peut
    0.42
     może
    0.41
     maybe
    0.40
     могÑĥÑĤ
    0.39
    Activations Density 0.018%

    No Known Activations