© 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 49152
    9. 8-RES_FS49152-JB
    11. 26793
    Prev
    Next
    INDEX
    Explanations

    references to geometric concepts and structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    +0.14
    0.6%
    235
    +0.12
    0.5%
    523
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.01
    235
    +0.12
    0.01
    326
    +0.10
    0.01
    Negative Logits
    ership
    -0.93
    ouse
    -0.74
    oice
    -0.72
    fleet
    -0.72
    ocked
    -0.71
    ousing
    -0.70
    ammy
    -0.70
    rior
    -0.69
    iken
    -0.68
    osi
    -0.67
    POSITIVE LOGITS
     shapes
    1.10
     symmetry
    0.99
     diagrams
    0.98
    allic
    0.97
     drawings
    0.95
     geometry
    0.87
     mesh
    0.86
     sculptures
    0.86
     deform
    0.85
     fract
    0.84
    Activations Density 0.007%

    No Known Activations