© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 3318
    Prev
    Next
    INDEX
    Explanations

    mentions of criminal charges and sentences related to computer fraud

    oai_token-act-pair · gpt-3.5-turbo
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    640
    +0.28
    1.4%
    160
    +0.19
    0.9%
    314
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.28
    0.07
    160
    +0.19
    0.07
    6
    +0.15
    0.06
    Negative Logits
     smugglers
    -0.89
     pse
    -0.86
     speeding
    -0.82
     treaty
    -0.81
     accent
    -0.80
     bloc
    -0.80
     trump
    -0.80
     ridic
    -0.79
     hyp
    -0.79
     polic
    -0.77
    POSITIVE LOGITS
    Same
    1.51
    Inv
    1.40
    Topic
    1.39
    Used
    1.38
    Contact
    1.37
    Return
    1.36
    Premium
    1.35
    Nothing
    1.35
    Use
    1.35
    Very
    1.33
    Activations Density 0.117%

    No Known Activations