© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 63285
    Prev
    Next
    INDEX
    Explanations

    punctuations and sentence structures

    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
    mai
    -0.16
    mate
    -0.16
    emente
    -0.15
    apis
    -0.15
    meter
    -0.15
    amak
    -0.15
    ãĥŁ
    -0.14
     ettir
    -0.14
     Mam
    -0.14
    upert
    -0.14
    POSITIVE LOGITS
     Mor
    1.41
    Mor
    1.41
     mor
    1.36
     MOR
    1.30
     Morris
    1.20
    mor
    1.18
     ÐľÐ¾ÑĢ
    1.01
     morph
    0.99
     Morph
    0.95
     morality
    0.94
    Activations Density 0.011%

    No Known Activations