© 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. 1270
    Prev
    Next
    INDEX
    Explanations

    contractions of the words "would" and "had"

    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
    712
    +0.11
    0.5%
    8
    +0.10
    0.5%
    566
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    566
    +0.11
    0.03
    661
    +0.10
    0.02
    354
    +0.10
    0.02
    Negative Logits
    PI
    -0.67
    aders
    -0.66
     Gutenberg
    -0.63
     Hels
    -0.63
    IDA
    -0.63
    士
    -0.59
    strings
    -0.59
     MODULE
    -0.59
    vich
    -0.58
     Compass
    -0.58
    POSITIVE LOGITS
     been
    1.07
     gotten
    1.05
     gotta
    1.01
     gone
    0.93
    been
    0.84
    hran
    0.80
     Been
    0.80
     gladly
    0.78
     got
    0.76
     ridden
    0.75
    Activations Density 0.016%

    No Known Activations