© 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 EvalsExportsinterp-engineNEWAPI 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. 22286
    Prev
    Next
    INDEX
    Explanations

    references to investigations, particularly those involving political figures and allegations

    oai_token-act-pair · gpt-4o-miniTriggered 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
    174
    +0.10
    0.4%
    362
    +0.09
    0.4%
    732
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.10
    0.07
    174
    +0.09
    0.04
    291
    +0.09
    0.05
    Negative Logits
    anguage
    -0.81
    izons
    -0.79
    iku
    -0.77
    icons
    -0.76
     partName
    -0.74
    erd
    -0.74
    FIG
    -0.74
    JV
    -0.73
     âĺ
    -0.71
    erm
    -0.70
    POSITIVE LOGITS
     misconduct
    1.28
     mishand
    1.25
     alleged
    1.19
     wrongdoing
    1.14
     allegations
    1.14
     botched
    1.13
     dealings
    1.12
     crimes
    1.12
     corruption
    1.12
     finances
    1.11
    Activations Density 0.512%

    No Known Activations