© 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. 3117
    Prev
    Next
    INDEX
    Explanations

    instructions or procedures

    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
    678
    +0.14
    0.6%
    658
    +0.09
    0.4%
    690
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    678
    +0.14
    0.06
    35
    +0.09
    0.03
    594
    +0.09
    0.03
    Negative Logits
    vernment
    -0.72
    女
    -0.71
     tickets
    -0.70
     successors
    -0.70
    ticket
    -0.68
    ijah
    -0.67
    alien
    -0.66
    IRD
    -0.65
     emblem
    -0.65
    isphere
    -0.65
    POSITIVE LOGITS
     pH
    1.45
     cooled
    1.36
     drying
    1.27
     fermentation
    1.25
     dried
    1.23
     electroly
    1.21
     boiling
    1.21
     rinse
    1.20
     soak
    1.19
     curing
    1.18
    Activations Density 0.288%

    No Known Activations