© 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. 15508
    Prev
    Next
    INDEX
    Explanations

    words related to future plans or goals

    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
    55
    +0.15
    0.7%
    132
    +0.13
    0.6%
    285
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.15
    0.04
    132
    +0.13
    0.04
    513
    +0.10
    0.03
    Negative Logits
    icipated
    -0.67
    idia
    -0.63
    inarily
    -0.61
     Hammond
    -0.57
     takedown
    -0.56
    hyde
    -0.56
    cu
    -0.55
    enance
    -0.54
    essential
    -0.54
     Champions
    -0.54
    POSITIVE LOGITS
    oths
    0.82
     much
    0.75
    ħĭ
    0.74
     afraid
    0.71
     guiActiveUn
    0.71
    oji
    0.71
    çĭ
    0.69
     :)
    0.69
     far
    0.69
    !.
    0.67
    Activations Density 0.031%

    No Known Activations