© 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 EvalsExportsinterp-engineNEWAPI 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. 18371
    Prev
    Next
    INDEX
    Explanations

    prepositions followed by specific nouns or phrases

    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
    314
    +0.12
    0.5%
    569
    +0.11
    0.5%
    386
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    569
    +0.12
    0.02
    123
    +0.11
    0.02
    750
    +0.11
    0.01
    Negative Logits
     goodbye
    -0.71
     remedies
    -0.68
     aloud
    -0.67
     replacements
    -0.65
     descendants
    -0.64
    GBT
    -0.64
    chairs
    -0.63
     fortun
    -0.63
    tons
    -0.61
     invitations
    -0.61
    POSITIVE LOGITS
     Breach
    0.85
     traps
    0.84
     trap
    0.83
    oppers
    0.67
     surprise
    0.66
    arc
    0.66
    unders
    0.64
     mistake
    0.63
    seys
    0.63
    Cross
    0.63
    Activations Density 0.065%

    No Known Activations