© 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. 54730
    Prev
    Next
    INDEX
    Explanations

    phrases indicating negative experiences or complaints

    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
    ndon
    -0.13
    lej
    -0.13
    аÑĤив
    -0.12
    isas
    -0.12
    argar
    -0.12
    .ManyToMany
    -0.12
    idebar
    -0.11
    tember
    -0.11
    ÏĦια
    -0.11
    âķIJâķIJ
    -0.11
    POSITIVE LOGITS
     now
    1.48
    now
    1.19
     Now
    1.11
    Now
    1.02
     NOW
    1.01
     ahora
    0.98
     ÑĤепеÑĢÑĮ
    0.96
    -now
    0.92
     maintenant
    0.92
    _now
    0.90
    Activations Density 0.099%

    No Known Activations