© 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. 29483
    Prev
    Next
    INDEX
    Explanations

    numerical representations of polling data and support levels

    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
    Ñģклад
    -0.19
     gratuitement
    -0.15
    addtogroup
    -0.14
     دÛĮگرÛĮ
    -0.14
    âĦĸâĦĸ
    -0.14
    zens
    -0.14
    ÐIJÑĢÑħÑĸв
    -0.14
    ãģ°ãģĭãĤĬ
    -0.13
     numbering
    -0.13
     ÑĹ
    -0.13
    POSITIVE LOGITS
     respectively
    0.23
    å·¦åı³
    0.18
    à¹Ģศษ
    0.17
     overall
    0.16
     ap
    0.15
    eking
    0.14
    itre
    0.13
    etre
    0.13
    overall
    0.13
     minimum
    0.13
    Activations Density 0.151%

    No Known Activations