© 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. 13253
    Prev
    Next
    INDEX
    Explanations

    contrasting statements or refutations

    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
    inite
    -0.18
    ilet
    -0.15
    razy
    -0.15
    Invariant
    -0.14
    æļ
    -0.14
    à¹īวย
    -0.13
    ึà¸ĩ
    -0.13
    å°ļ
    -0.13
    osite
    -0.13
    بÙĬØ©
    -0.13
    POSITIVE LOGITS
     actually
    0.47
     reality
    0.46
     Actually
    0.42
    actually
    0.39
     Reality
    0.38
    Actually
    0.38
     actual
    0.38
    Reality
    0.37
    å®ŀéĻħ
    0.34
     realidad
    0.34
    Activations Density 0.075%

    No Known Activations