© 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. 3910
    Prev
    Next
    INDEX
    Explanations

    keywords related to specific events, updates, or changes happening within a certain context such as sports, economics, or politics

    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
    538
    +0.10
    0.4%
    282
    +0.09
    0.4%
    128
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    155
    +0.10
    0.03
    538
    +0.09
    0.02
    495
    +0.09
    0.02
    Negative Logits
    iary
    -0.92
    bern
    -0.76
    pless
    -0.75
    iaries
    -0.75
     Flavoring
    -0.68
    asio
    -0.68
    resents
    -0.66
    gee
    -0.64
    aye
    -0.64
    lication
    -0.63
    POSITIVE LOGITS
     transpired
    1.04
     happened
    1.01
     happen
    1.00
     happening
    0.89
     unfolded
    0.88
     unfold
    0.87
     pertaining
    0.86
     escalated
    0.86
     escalate
    0.82
     Happ
    0.78
    Activations Density 0.037%

    No Known Activations