© 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. 22608
    Prev
    Next
    INDEX
    Explanations

    occurrences of the pronoun "I"

    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
    373
    +0.23
    1.1%
    76
    +0.10
    0.5%
    192
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.23
    -0.01
    232
    +0.10
    0.05
    659
    +0.09
    0.05
    Negative Logits
    advertisement
    -0.63
     accomp
    -0.61
     Arist
    -0.60
    RAFT
    -0.60
    ÄŁ
    -0.57
    iaries
    -0.56
    ymes
    -0.56
    _-
    -0.56
    ONSORED
    -0.55
    ambers
    -0.55
    POSITIVE LOGITS
    isha
    0.73
    weather
    0.68
    boys
    0.66
    sex
    0.65
    azi
    0.64
    oya
    0.64
    coe
    0.64
    key
    0.63
    way
    0.63
    elle
    0.62
    Activations Density 0.114%

    No Known Activations