© 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 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. 5957
    Prev
    Next
    INDEX
    Explanations

    references to specific technology or telecommunication companies

    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
    +0.17
    0.8%
    756
    +0.09
    0.4%
    641
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.17
    -0.01
    701
    +0.09
    0.02
    311
    +0.09
    0.01
    Negative Logits
     Welch
    -0.68
     Jordan
    -0.62
     roots
    -0.59
     Loll
    -0.58
     Burnett
    -0.58
     AMA
    -0.57
     chrom
    -0.57
     Garr
    -0.57
     regards
    -0.57
     Fargo
    -0.56
    POSITIVE LOGITS
    ield
    1.14
    ruit
    1.07
    riend
    1.05
    iction
    1.03
    requency
    1.02
    actor
    0.99
    actory
    0.98
    raction
    0.98
    ascist
    0.94
    ront
    0.94
    Activations Density 0.044%

    No Known Activations