© 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. 15335
    Prev
    Next
    INDEX
    Explanations

    phrases related to action and completion

    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%
    685
    +0.13
    0.6%
    197
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    553
    +0.14
    0.04
    685
    +0.13
    0.03
    125
    +0.11
    0.03
    Negative Logits
    ãĥĥãĤ¯
    -0.68
     DRAG
    -0.65
    ourses
    -0.64
    anton
    -0.62
    NEWS
    -0.59
    anty
    -0.57
    uri
    -0.56
     Fatal
    -0.55
     Patron
    -0.54
    Song
    -0.54
    POSITIVE LOGITS
     cheaply
    0.94
     aloud
    0.83
     somehow
    0.80
     anyway
    0.79
     again
    0.78
     altogether
    0.77
     individually
    0.76
     manually
    0.76
     promptly
    0.76
     yourself
    0.75
    Activations Density 0.116%

    No Known Activations