© 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. 1506
    Prev
    Next
    INDEX
    Explanations

    written statements describing decisions or judgments based on personal beliefs or experiences

    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
    46
    +0.11
    0.5%
    617
    +0.11
    0.5%
    242
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    661
    +0.11
    0.03
    7
    +0.11
    0.04
    566
    +0.10
    0.04
    Negative Logits
    iffin
    -0.71
     Appears
    -0.70
    byn
    -0.69
     âĵĺ
    -0.67
    anova
    -0.65
    lez
    -0.65
     NX
    -0.65
    xit
    -0.64
    rection
    -0.64
    wikipedia
    -0.63
    POSITIVE LOGITS
     outwe
    0.80
     cheap
    0.74
     outnumbered
    0.73
     darn
    0.73
     pree
    0.73
     cheaper
    0.72
     menstru
    0.70
     cumbers
    0.68
     inconvenient
    0.67
     too
    0.67
    Activations Density 0.422%

    No Known Activations