© 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. 3045
    Prev
    Next
    INDEX
    Explanations

    legislation-related terms and political figures' names

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    350
    +0.12
    0.6%
    398
    +0.10
    0.5%
    31
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    690
    +0.12
    0.04
    398
    +0.10
    0.04
    300
    +0.10
    0.05
    Negative Logits
     pearl
    -0.69
     Himal
    -0.69
     Divinity
    -0.69
     Atlantis
    -0.69
     outage
    -0.69
     civilisation
    -0.67
     Celtic
    -0.66
     sensed
    -0.66
    Pont
    -0.66
     paranormal
    -0.66
    POSITIVE LOGITS
    ospons
    1.46
    osponsors
    1.38
     amend
    1.24
     amended
    1.21
     repeal
    1.17
     legislation
    1.11
     amendments
    1.10
     repealed
    1.10
     proposed
    1.09
     decriminal
    1.07
    Activations Density 0.266%

    No Known Activations