© 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. 24466
    Prev
    Next
    INDEX
    Explanations

    elements related to visual and thematic qualities in media and art

    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
    373
    +0.18
    0.9%
    163
    +0.11
    0.5%
    367
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    760
    +0.18
    0.03
    576
    +0.11
    0.02
    163
    +0.10
    0.03
    Negative Logits
    lection
    -0.88
    ember
    -0.81
    ertation
    -0.80
    anders
    -0.77
    onductor
    -0.77
    ãĤ¼ãĤ¦ãĤ¹
    -0.73
    azine
    -0.73
    ahime
    -0.73
    ç¥ŀ
    -0.72
     æľ
    -0.72
    POSITIVE LOGITS
     phony
    0.78
     Freak
    0.73
     freak
    0.72
     goofy
    0.71
     kinda
    0.70
     mega
    0.70
     goof
    0.70
     sill
    0.70
     mish
    0.69
     poke
    0.69
    Activations Density 0.102%

    No Known Activations