© 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. 63199
    Prev
    Next
    INDEX
    Explanations

    modal verbs indicating ability or possibility

    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
    овиÑĩ
    -0.16
    ufe
    -0.16
    edom
    -0.16
    lette
    -0.15
    cano
    -0.15
    uling
    -0.15
    ural
    -0.14
    ìŀĶ
    -0.14
    ë°ķ
    -0.13
    äll
    -0.13
    POSITIVE LOGITS
    't
    0.43
    ’t
    0.41
    berra
    0.24
    tered
    0.24
    ister
    0.23
    opies
    0.23
     not
    0.22
    opy
    0.21
     only
    0.21
    ibal
    0.20
    Activations Density 0.257%

    No Known Activations