© 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. 23821
    Prev
    Next
    INDEX
    Explanations

    descriptions of processes related to observations and measurements

    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
    vÄĽt
    -0.18
    laden
    -0.15
     Fucking
    -0.14
    à¸ľ
    -0.14
     blowjob
    -0.14
     abst
    -0.14
    567
    -0.14
    asso
    -0.13
     fucking
    -0.13
     moh
    -0.13
    POSITIVE LOGITS
    BigInteger
    0.16
    ân
    0.15
     overall
    0.15
    çIJ
    0.15
     billions
    0.15
    .forChild
    0.14
     tens
    0.14
    ุà¸ķ
    0.14
    ãģĺ
    0.14
    -envelope
    0.14
    Activations Density 0.342%

    No Known Activations