© 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. 15551
    Prev
    Next
    INDEX
    Explanations

    informal language indicating hurried or rushed activity

    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
    +0.13
    0.6%
    271
    +0.12
    0.6%
    580
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.13
    0.02
    326
    +0.12
    0.01
    70
    +0.09
    0.02
    Negative Logits
    hetical
    -0.81
    osate
    -0.70
    paying
    -0.66
    ucc
    -0.61
    cause
    -0.61
    ithe
    -0.60
    neutral
    -0.60
    negative
    -0.59
    Sharp
    -0.59
     Vict
    -0.59
    POSITIVE LOGITS
     scramble
    1.25
     scram
    0.96
     scrambling
    0.95
     scrambled
    0.90
     shuffle
    0.82
     frantically
    0.81
    berry
    0.78
     furiously
    0.73
    DAQ
    0.72
    weed
    0.71
    Activations Density 0.044%

    No Known Activations