© 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. 19607
    Prev
    Next
    INDEX
    Explanations

    proper names and historical terms

    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
    447
    +0.13
    0.6%
    659
    +0.11
    0.5%
    232
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.07
    232
    +0.11
    0.06
    284
    +0.10
    0.05
    Negative Logits
     affordability
    -0.82
     analytics
    -0.79
     robotics
    -0.75
     scoreboard
    -0.73
     broker
    -0.70
     sustainability
    -0.68
     minimized
    -0.68
     firewall
    -0.68
     muted
    -0.67
     academic
    -0.67
    POSITIVE LOGITS
    aceae
    1.15
    idae
    1.14
    Åį
    0.98
    ii
    0.95
    ubi
    0.95
    iae
    0.94
    irs
    0.93
    aurus
    0.92
    hai
    0.92
    bol
    0.91
    Activations Density 0.232%

    No Known Activations