© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 63489
    Prev
    Next
    INDEX
    Explanations

    references to autistic and non-autistic individuals in relation to their social behaviors and traits

    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
    ÏĦÏİ
    -0.15
    çĶļ
    -0.14
    ÑĢава
    -0.14
    omitempty
    -0.14
     заÑģÑĤ
    -0.13
    пÑĢимеÑĢ
    -0.13
    .sy
    -0.13
    oard
    -0.13
    leston
    -0.13
    essim
    -0.13
    POSITIVE LOGITS
     non
    1.10
     Non
    0.94
    non
    0.93
    Non
    0.91
     NON
    0.88
    éĿŀ
    0.83
    -non
    0.82
    _non
    0.79
    .non
    0.76
    (non
    0.72
    Activations Density 0.029%

    No Known Activations