© 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. 24503
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "to be" in various forms and tenses

    oai_token-act-pair · gpt-4o-miniTriggered 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
    55
    +0.16
    0.8%
    481
    +0.12
    0.6%
    566
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.16
    0.09
    102
    +0.12
    0.08
    566
    +0.10
    0.07
    Negative Logits
    inav
    -0.78
     impulses
    -0.74
     congratulations
    -0.73
    worthiness
    -0.73
     congr
    -0.69
    isms
    -0.69
     defeats
    -0.69
     efficacy
    -0.66
     convictions
    -0.66
     conviction
    -0.66
    POSITIVE LOGITS
     situated
    1.17
     located
    1.15
     inhabited
    1.11
     dotted
    1.11
     flooded
    1.11
     overrun
    1.05
     populated
    1.04
     littered
    1.03
     housed
    1.03
     inund
    1.01
    Activations Density 0.259%

    No Known Activations