© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 9547
    Prev
    Next
    INDEX
    Explanations

    mentions of legal proceedings or litigation

    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
    571
    +0.10
    0.5%
    19
    +0.10
    0.5%
    566
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    571
    +0.10
    0.01
    159
    +0.10
    0.01
    252
    +0.09
    0.01
    Negative Logits
    Downloadha
    -0.76
    FactoryReloaded
    -0.70
    088
    -0.68
    CVE
    -0.68
    ä½ľ
    -0.66
     Immun
    -0.65
    REDACTED
    -0.65
    axter
    -0.64
    SPONSORED
    -0.63
    DonaldTrump
    -0.62
    POSITIVE LOGITS
    urgy
    1.32
    ters
    1.25
    urgical
    1.23
    igated
    1.17
    igating
    1.13
    igators
    1.12
    igious
    1.10
    igator
    1.09
    igation
    1.03
    eral
    1.01
    Activations Density 0.022%

    No Known Activations