© 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. 2093
    Prev
    Next
    INDEX
    Explanations

    information related to events, announcements, and specific dates

    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
    28
    +0.12
    0.5%
    486
    +0.11
    0.5%
    492
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.04
    28
    +0.11
    0.03
    5
    +0.10
    0.03
    Negative Logits
     bleach
    -0.86
    207
    -0.76
     Bleach
    -0.72
    ¡
    -0.71
     Goldberg
    -0.68
    Ble
    -0.65
    ł
    -0.65
    ĵĺ
    -0.63
    itz
    -0.63
    idth
    -0.62
    POSITIVE LOGITS
    m
    1.41
    M
    1.22
    mo
    1.12
    mic
    1.11
    mill
    1.09
    Ms
    1.08
    mt
    1.08
    MI
    1.07
    mob
    1.06
    mk
    1.06
    Activations Density 0.190%

    No Known Activations