© 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. 14907
    Prev
    Next
    INDEX
    Explanations

    positive or motivational words

    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
    447
    +0.32
    1.6%
    480
    +0.20
    0.9%
    481
    +0.18
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    266
    +0.32
    0.04
    464
    +0.20
    0.04
    480
    +0.18
    0.01
    Negative Logits
    Nich
    -0.74
    Bride
    -0.70
    ãĥ¼ãĥĨ
    -0.67
     Vaugh
    -0.66
    FN
    -0.64
    Materials
    -0.63
    Marie
    -0.63
    Maria
    -0.63
    Grade
    -0.62
    Safety
    -0.62
    POSITIVE LOGITS
     english
    0.81
     dont
    0.70
    xual
    0.66
     doesnt
    0.63
     sd
    0.63
    nt
    0.63
    ername
    0.60
     fucked
    0.59
     tho
    0.58
     flair
    0.58
    Activations Density 0.183%

    No Known Activations