© 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. 18541
    Prev
    Next
    INDEX
    Explanations

    comparisons using the word "like"

    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
    373
    +0.12
    0.6%
    583
    +0.10
    0.5%
    512
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    172
    +0.12
    0.03
    244
    +0.10
    0.02
    117
    +0.10
    0.02
    Negative Logits
    byn
    -0.80
    conservancy
    -0.67
    earance
    -0.64
    alez
    -0.64
    arcity
    -0.64
    itions
    -0.64
    alt
    -0.63
    oust
    -0.63
    edom
    -0.63
    izarre
    -0.63
    POSITIVE LOGITS
     crap
    0.98
    lier
    0.93
     shit
    0.84
     idiots
    0.73
     fools
    0.71
    liest
    0.68
     they
    0.67
     outsiders
    0.67
    lihood
    0.67
     THEY
    0.65
    Activations Density 0.032%

    No Known Activations