© 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. 16021
    Prev
    Next
    INDEX
    Explanations

    information and discussions related to politics and current events

    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
    679
    +0.21
    1.0%
    481
    +0.19
    0.9%
    430
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.21
    0.12
    679
    +0.19
    0.12
    430
    +0.13
    0.10
    Negative Logits
     reversible
    -0.88
     clutch
    -0.78
     wake
    -0.73
     applause
    -0.70
     awaited
    -0.69
     consolation
    -0.69
     motivational
    -0.69
     therapeutic
    -0.68
     chorus
    -0.67
     captcha
    -0.67
    POSITIVE LOGITS
     Therefore
    1.48
     Its
    1.40
     Hence
    1.37
     Consequently
    1.34
     Moreover
    1.28
     Accordingly
    1.25
     Thus
    1.18
     Yet
    1.17
     Presumably
    1.14
     Furthermore
    1.12
    Activations Density 0.508%

    No Known Activations