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

    Neuronpedia

    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 3072
    9. 8-RES_FS3072-JB
    11. 781
    Prev
    Next
    INDEX
    Explanations

    references to specific items or products along with their prices

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    523
    +0.10
    0.4%
    274
    +0.09
    0.4%
    289
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    289
    +0.10
    0.03
    232
    +0.09
    0.04
    326
    +0.09
    0.04
    Negative Logits
    enegger
    -0.74
    ICLE
    -0.72
    ral
    -0.72
    ãģ®éŃĶ
    -0.70
    bell
    -0.69
     pleas
    -0.69
    ãĥ£
    -0.64
    NEY
    -0.64
    terms
    -0.63
    æĦ
    -0.63
    POSITIVE LOGITS
    ideshow
    1.16
    ippery
    1.09
    udge
    1.09
    ips
    1.08
    ickers
    1.08
    ashes
    1.07
    anted
    1.06
    ugg
    1.05
    ashing
    1.04
    icker
    1.04
    Activations Density 0.484%

    No Known Activations