© 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. 30049
    Prev
    Next
    INDEX
    Explanations

    suffixes and prefixes commonly used in technical terms

    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
    ients
    -0.14
    ili
    -0.14
    curacy
    -0.13
    ÅĽÄĩ
    -0.13
    @Spring
    -0.13
    iline
    -0.13
    ksam
    -0.12
    ilece
    -0.12
    ilecek
    -0.12
    eness
    -0.12
    POSITIVE LOGITS
     Aires
    0.14
     (*((
    0.14
    [email
    0.14
     ActionTypes
    0.13
    HING
    0.13
    _ALWAYS
    0.13
    ï¼Ĭ
    0.13
    ıi
    0.13
    ãİ
    0.13
    _Lean
    0.13
    Activations Density 0.005%

    No Known Activations