© 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. 4001
    Prev
    Next
    INDEX
    Explanations

    personal pronouns indicating community or group affiliation

    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
    492
    +0.11
    0.5%
    196
    +0.11
    0.5%
    571
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    502
    +0.11
    0.04
    492
    +0.11
    0.04
    139
    +0.10
    0.04
    Negative Logits
     Majority
    -0.79
     MV
    -0.67
     srfAttach
    -0.65
     Debor
    -0.64
     Petraeus
    -0.62
     Soda
    -0.58
     Acceler
    -0.57
     Sop
    -0.56
     Nora
    -0.56
     Cla
    -0.56
    POSITIVE LOGITS
     encount
    0.96
     deems
    0.90
    've
    0.88
     encountered
    0.83
    needed
    0.83
    'd
    0.80
     ever
    0.78
     feas
    0.77
     touched
    0.76
     deem
    0.76
    Activations Density 0.102%

    No Known Activations