© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 8-RES-JB
    11. 2051
    Prev
    Next
    INDEX
    Explanations

    references to the phrase "Attack on"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of attacks and the term "Attack" across 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
    YC
    -0.80
    zl
    -0.78
    ãĤ©
    -0.75
    inders
    -0.67
    mberg
    -0.63
    姫
    -0.63
    inder
    -0.62
    theless
    -0.60
    teness
    -0.60
    essage
    -0.59
    POSITIVE LOGITS
     against
    0.87
     vector
    0.84
    CVE
    0.77
    ivated
    0.76
    iveness
    0.75
     vectors
    0.74
    ivation
    0.74
    attack
    0.73
     attacks
    0.73
     [+]
    0.72
    Activations Density 0.061%

    No Known Activations