© 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. 24351
    Prev
    Next
    INDEX
    Explanations

    words related to names and specific identifiers

    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
    296
    +0.11
    0.5%
    544
    +0.11
    0.5%
    91
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.11
    0.04
    659
    +0.11
    0.03
    760
    +0.10
    0.02
    Negative Logits
    ãĤ¬
    -0.77
     ESA
    -0.75
     DIRECT
    -0.73
     Manit
    -0.73
     arrang
    -0.71
     76
    -0.71
     CAP
    -0.71
     helicop
    -0.70
     802
    -0.70
     Cort
    -0.69
    POSITIVE LOGITS
    y
    1.61
    ys
    1.40
    yon
    1.35
    yles
    1.34
    yl
    1.29
    yx
    1.29
    yg
    1.29
    Y
    1.27
    ym
    1.25
    yan
    1.24
    Activations Density 0.108%

    No Known Activations