© 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. 2966
    Prev
    Next
    INDEX
    Explanations

    information related to government actions and legislation

    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
    87
    +0.23
    1.2%
    373
    +0.12
    0.6%
    712
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    516
    +0.23
    0.05
    575
    +0.12
    0.07
    329
    +0.10
    0.05
    Negative Logits
    çĶ°
    -0.71
    yssey
    -0.63
     largeDownload
    -0.62
     realise
    -0.61
     Jaguar
    -0.58
     Guru
    -0.58
     apologise
    -0.58
    Picture
    -0.57
    DragonMagazine
    -0.57
    NUM
    -0.57
    POSITIVE LOGITS
     lawfully
    1.03
     lawful
    0.99
     unlawfully
    0.98
     federally
    0.96
     unlawful
    0.94
     unconstitutional
    0.91
     constitutionally
    0.85
     infringing
    0.83
     violates
    0.82
     infring
    0.81
    Activations Density 0.838%

    No Known Activations