© 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. 25627
    Prev
    Next
    INDEX
    Explanations

    phrases related to statements and claims, particularly involving assessments of truthfulness or accuracy

    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
    anz
    -0.19
    éĻµ
    -0.15
    ago
    -0.14
    airo
    -0.14
     Jordan
    -0.14
    è«ĸ
    -0.14
    960
    -0.14
    ông
    -0.14
    za
    -0.14
    åĢĴ
    -0.14
    POSITIVE LOGITS
     underst
    0.43
     mild
    0.26
     charitable
    0.24
    under
    0.23
    Under
    0.22
     Mild
    0.22
     gentle
    0.22
    UNDER
    0.21
    -under
    0.21
    .PUT
    0.21
    Activations Density 0.013%

    No Known Activations