© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 16835
    Prev
    Next
    INDEX
    Explanations

    references to significant events or accomplishments in sports, particularly in hockey

    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
    +0.15
    0.7%
    391
    +0.10
    0.5%
    505
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    422
    +0.15
    0.03
    711
    +0.10
    0.02
    290
    +0.10
    0.03
    Negative Logits
     Canaver
    -0.70
     PLA
    -0.70
     Squid
    -0.70
    cgi
    -0.70
    cms
    -0.70
    terday
    -0.70
    apa
    -0.68
    OPA
    -0.67
    echo
    -0.67
     ITS
    -0.65
    POSITIVE LOGITS
     stint
    1.36
     appearances
    1.17
     career
    1.11
     accol
    1.10
     appearance
    1.03
     outing
    1.00
     pedigree
    0.95
     careers
    0.91
     tenure
    0.89
     qualification
    0.88
    Activations Density 0.149%

    No Known Activations