© 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 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. 17871
    Prev
    Next
    INDEX
    Explanations

    mentions of names, titles, and organizations

    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
    +0.16
    0.7%
    592
    +0.15
    0.7%
    684
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    443
    +0.16
    0.04
    659
    +0.15
    0.05
    83
    +0.11
    0.03
    Negative Logits
     WTO
    -0.77
    ndum
    -0.76
     uninsured
    -0.73
     leaflets
    -0.72
    Vote
    -0.71
    Statistics
    -0.71
     Labrador
    -0.71
     precincts
    -0.70
     respondent
    -0.70
    daq
    -0.70
    POSITIVE LOGITS
     creatively
    1.17
     Productions
    1.14
     Studios
    1.10
     collaborated
    1.06
     collaborator
    1.06
     Studio
    1.00
     Animation
    0.99
     composer
    0.98
     Compos
    0.97
     collaborating
    0.97
    Activations Density 0.523%

    No Known Activations