© 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. 20788
    Prev
    Next
    INDEX
    Explanations

    locations or physical structures

    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
    481
    +0.21
    1.0%
    291
    +0.12
    0.5%
    413
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    564
    +0.21
    0.04
    291
    +0.12
    0.04
    221
    +0.12
    0.03
    Negative Logits
    ¥
    -0.79
    İĭ
    -0.76
    Real
    -0.72
    Week
    -0.70
    °
    -0.70
     Childhood
    -0.70
    Money
    -0.69
    Ļ½
    -0.69
    Beat
    -0.68
    rosso
    -0.68
    POSITIVE LOGITS
     protr
    1.09
     liner
    1.00
     pouch
    1.00
     shaped
    0.98
     enclosure
    0.98
     containing
    0.98
     cutter
    0.95
     necklace
    0.93
     casing
    0.93
    hole
    0.92
    Activations Density 0.206%

    No Known Activations