© 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. 24570
    Prev
    Next
    INDEX
    Explanations

    references to specific geographic locations and their associated events or entities

    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
    48
    +0.11
    0.5%
    219
    +0.10
    0.4%
    388
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    479
    +0.11
    0.03
    6
    +0.10
    0.04
    48
    +0.09
    0.03
    Negative Logits
    prus
    -1.02
    angel
    -0.90
    ija
    -0.89
    organ
    -0.88
    Columb
    -0.87
    ascus
    -0.87
    rug
    -0.86
    bitcoin
    -0.85
    sung
    -0.85
    silver
    -0.85
    POSITIVE LOGITS
     M
    1.26
     C
    1.17
     N
    1.15
     R
    1.14
     S
    1.13
     L
    1.13
     G
    1.11
     Y
    1.09
     H
    1.08
     E
    1.08
    Activations Density 0.180%

    No Known Activations