© 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. 8588
    Prev
    Next
    INDEX
    Explanations

    names that include the term "Mc" with a specific activation pattern

    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
    659
    +0.20
    0.9%
    575
    +0.10
    0.5%
    508
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.20
    0.04
    508
    +0.10
    0.02
    232
    +0.10
    0.02
    Negative Logits
     Gemini
    -0.76
     Guilty
    -0.71
    DragonMagazine
    -0.69
     firsthand
    -0.67
     prevail
    -0.65
     unprotected
    -0.65
     paranormal
    -0.64
     CBI
    -0.63
     Bulgarian
    -0.62
     Acts
    -0.62
    POSITIVE LOGITS
    Afee
    1.41
    intosh
    1.34
    Leod
    1.31
    donald
    1.27
    lean
    1.15
    artney
    1.12
    Lean
    1.12
    Bride
    1.04
    gee
    1.01
    arthy
    0.99
    Activations Density 0.009%

    No Known Activations