© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 3165
    Prev
    Next
    INDEX
    Explanations

    mentions of locations, especially cities and states

    oai_token-act-pair · gpt-3.5-turbo
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    380
    +0.10
    0.4%
    314
    +0.09
    0.4%
    55
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    140
    +0.10
    0.03
    425
    +0.09
    0.03
    587
    +0.09
    0.03
    Negative Logits
    hift
    -0.83
    obin
    -0.64
    20439
    -0.64
    ariat
    -0.63
    erity
    -0.63
    hide
    -0.63
    iral
    -0.62
    isy
    -0.61
    Afee
    -0.60
    lash
    -0.59
    POSITIVE LOGITS
     where
    0.79
     birthplace
    0.76
     USA
    0.73
     circa
    0.71
     respectively
    0.68
     southwest
    0.68
     Calif
    0.68
     northwest
    0.65
    today
    0.64
     shortly
    0.63
    Activations Density 0.080%

    No Known Activations