© 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. 18748
    Prev
    Next
    INDEX
    Explanations

    terms related to legal actions or implications, such as 'pro', 'jail', 'resign', and 'prison'

    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
    +0.34
    1.7%
    447
    +0.30
    1.5%
    665
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.34
    0.09
    762
    +0.30
    0.07
    761
    +0.11
    0.06
    Negative Logits
     Nielsen
    -0.54
     NFC
    -0.52
     Soph
    -0.51
     Gem
    -0.51
     Bulg
    -0.50
     brow
    -0.50
     Surface
    -0.49
     gulf
    -0.49
     Fulton
    -0.48
    idth
    -0.48
    POSITIVE LOGITS
     sooner
    1.15
     anytime
    1.04
     ASAP
    1.00
     voluntarily
    0.98
     altogether
    0.96
     someday
    0.96
     anyway
    0.95
     anyways
    0.86
     unless
    0.86
     because
    0.82
    Activations Density 0.501%

    No Known Activations