© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 15723
    Prev
    Next
    INDEX
    Explanations

    terms related to statistical changes, specifically increases and decreases

    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
    èªĮ
    -0.15
    neh
    -0.14
    à¸Ńà¸ļ
    -0.14
    ween
    -0.14
    adder
    -0.14
     ReturnValue
    -0.14
    nen
    -0.13
    Ľi
    -0.13
    AndPassword
    -0.13
    ìļ´ëį°
    -0.13
    POSITIVE LOGITS
     in
    0.46
     în
    0.24
    åľ¨
    0.24
    à¹ĥà¸Ļ
    0.22
     ÙģÙĬ
    0.21
    à¹ĥà¸Ļà¸ģาร
    0.20
     ÏĥÏĦην
    0.19
     dalam
    0.18
     trong
    0.18
    à¹ĥà¸Ļà¸Ĭ
    0.17
    Activations Density 0.111%

    No Known Activations