© 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. 22622
    Prev
    Next
    INDEX
    Explanations

    elements related to numerical data or statistics, particularly in the context of ratings or rankings

    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
    2
    +0.13
    0.6%
    189
    +0.12
    0.5%
    247
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    310
    +0.13
    0.04
    220
    +0.12
    0.04
    260
    +0.11
    0.04
    Negative Logits
    ophon
    -0.81
    paio
    -0.73
    ĸļ
    -0.70
    yle
    -0.68
    sein
    -0.68
    phis
    -0.66
    ei
    -0.64
    chwitz
    -0.63
    akening
    -0.62
    icago
    -0.62
    POSITIVE LOGITS
    th
    0.98
    00
    0.91
    isher
    0.90
    ishers
    0.89
    %-
    0.88
    60
    0.88
    ishing
    0.87
    %:
    0.87
    %
    0.86
    50
    0.83
    Activations Density 0.083%

    No Known Activations