Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 19624
    Prev
    Next
    INDEX
    Explanations

    phrases related to denial or rejection

    oai_token-act-pair · gpt-3.5-turbo

    instances of denial or refusal in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    oiler
    -0.86
    ARCH
    -0.74
    enegger
    -0.73
    Ec
    -0.71
    âĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪ
    -0.70
    INGTON
    -0.68
    andals
    -0.67
    è¦ļéĨĴ
    -0.67
    rim
    -0.66
    arnaev
    -0.66
    POSITIVE LOGITS
     access
    0.83
    ļéĨĴ
    0.82
    afe
    0.79
     gratification
    0.78
     entry
    0.77
     parole
    0.74
    lement
    0.71
     refunds
    0.69
     permission
    0.68
    ilege
    0.67
    Activations Density 0.061%

    No Known Activations