© 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. 17266
    Prev
    Next
    INDEX
    Explanations

    names of specific people or characters

    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
    659
    +0.21
    1.0%
    36
    +0.16
    0.7%
    314
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.21
    0.08
    36
    +0.16
    0.06
    213
    +0.11
    0.04
    Negative Logits
     unrestricted
    -0.72
     err
    -0.68
     fret
    -0.67
    tainment
    -0.66
     CPR
    -0.64
     unsupported
    -0.64
     techno
    -0.64
     dealing
    -0.63
     nic
    -0.63
     miracles
    -0.61
    POSITIVE LOGITS
    ixty
    1.41
    ustain
    1.23
    ESSION
    1.22
    esame
    1.20
    ometime
    1.18
    igm
    1.18
    udden
    1.17
    UD
    1.17
    IX
    1.16
    ixties
    1.14
    Activations Density 0.044%

    No Known Activations