© 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. 13041
    Prev
    Next
    INDEX
    Explanations

    occurrences of verbs indicating perception or state

    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
    xea
    -0.17
    onn
    -0.15
    le
    -0.15
     fewer
    -0.15
    None
    -0.14
     NONE
    -0.14
    mts
    -0.14
    _stub
    -0.13
    oga
    -0.13
    Only
    -0.13
    POSITIVE LOGITS
     no
    0.91
    no
    0.57
    .no
    0.50
    	no
    0.49
    _no
    0.46
    ,no
    0.46
     NO
    0.45
     absolutely
    0.41
    =no
    0.39
    -no
    0.39
    Activations Density 0.023%

    No Known Activations