© 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. 3762
    Prev
    Next
    INDEX
    Explanations

    verbs related to tasks, achievements, and commands in various contexts

    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
    481
    +0.14
    0.6%
    247
    +0.11
    0.5%
    160
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.08
    247
    +0.11
    0.06
    575
    +0.11
    0.07
    Negative Logits
    å°Ĩ
    -0.73
    ombs
    -0.67
    MET
    -0.67
    ategor
    -0.62
    ANN
    -0.60
    ILLE
    -0.60
    ortium
    -0.58
    itored
    -0.57
    iliated
    -0.57
    NK
    -0.57
    POSITIVE LOGITS
     itself
    1.11
    iest
    1.08
    osphere
    0.97
    ifice
    0.91
     quo
    0.88
     thereof
    0.85
    forts
    0.80
     of
    0.79
     unfolding
    0.78
    iveness
    0.77
    Activations Density 0.410%

    No Known Activations