© 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. 8995
    Prev
    Next
    INDEX
    Explanations

    references to specific research firms or research-related terms

    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
    679
    +0.12
    0.5%
    765
    +0.11
    0.5%
    713
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    235
    +0.12
    0.04
    679
    +0.11
    0.05
    219
    +0.10
    0.04
    Negative Logits
    Interstitial
    -0.71
    Control
    -0.70
     transporter
    -0.67
    Reply
    -0.64
     retaliate
    -0.64
     Leilan
    -0.63
     uncond
    -0.62
     Remem
    -0.62
    Î
    -0.62
    ã
    -0.61
    POSITIVE LOGITS
     polled
    1.09
     firm
    0.90
     analyst
    0.89
     methodology
    0.85
     consultancy
    0.82
     surveyed
    0.82
    verages
    0.80
     commissioned
    0.78
    gorithm
    0.77
     analyzed
    0.77
    Activations Density 0.170%

    No Known Activations