© 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. 6026
    Prev
    Next
    INDEX
    Explanations

    scientific and analytical terms used in discussions regarding various topics, including software, market analysis, and game development

    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
    36
    +0.21
    1.0%
    373
    +0.16
    0.7%
    640
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.21
    0.10
    159
    +0.16
    0.04
    250
    +0.09
    0.04
    Negative Logits
     outfield
    -0.65
     Edgar
    -0.63
     unaccompanied
    -0.62
    baugh
    -0.61
     tyres
    -0.61
     honors
    -0.61
     embell
    -0.61
     buttons
    -0.60
    overs
    -0.60
     Orioles
    -0.59
    POSITIVE LOGITS
     nutshell
    1.13
     glance
    1.02
     lot
    0.78
     glimpse
    0.78
    roma
    0.77
    ircraft
    0.76
     few
    0.76
    verages
    0.73
     curs
    0.72
     sense
    0.72
    Activations Density 0.129%

    No Known Activations