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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 12518
    Prev
    Next
    INDEX
    Explanations

    words related to vulnerability and security concerns, such as 'vulnerable' and 'vulnerability'

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    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

    Neuron Alignment
    Index
    Value
    % of L₁
    288
    +0.16
    0.8%
    496
    +0.13
    0.6%
    447
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    417
    +0.16
    0.02
    151
    +0.13
    0.02
    161
    +0.13
    0.01
    Negative Logits
     Collider
    -0.75
    zos
    -0.69
    inth
    -0.68
    atorium
    -0.67
    ateur
    -0.67
    rollers
    -0.65
    OTO
    -0.64
    Ħ¢
    -0.63
    ibal
    -0.62
    raining
    -0.62
    POSITIVE LOGITS
     vulnerable
    1.03
     vulner
    0.87
    ulnerable
    0.86
     vulnerability
    0.82
     vulnerabilities
    0.81
    ansas
    0.78
     weaknesses
    0.73
     exploited
    0.69
    ulnerability
    0.69
     adolesc
    0.69
    Activations Density 0.027%

    No Known Activations