© 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. 24206
    Prev
    Next
    INDEX
    Explanations

    various special characters and symbols within the text

    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
    447
    +0.24
    1.1%
    747
    +0.10
    0.4%
    437
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    107
    +0.24
    0.03
    103
    +0.10
    0.03
    284
    +0.09
    0.02
    Negative Logits
    espie
    -0.91
     skelet
    -0.80
     Quadro
    -0.78
     Dill
    -0.74
    enegger
    -0.73
     Jenkins
    -0.72
     PD
    -0.70
     Brennan
    -0.69
     Dunn
    -0.68
    ttes
    -0.67
    POSITIVE LOGITS
    ķ
    1.21
    ĺ
    1.18
    ł
    1.16
    Ķ
    1.14
    ¡
    1.12
    Ī
    1.12
    Ĩ
    1.11
    ľ
    1.10
    ħ
    1.10
    Ĵ
    1.07
    Activations Density 0.035%

    No Known Activations