© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 13251
    Prev
    Next
    INDEX
    Explanations

    sentences related to crimes or violent incidents

    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
    102
    +0.15
    0.7%
    55
    +0.12
    0.6%
    566
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.15
    0.08
    55
    +0.12
    0.07
    566
    +0.12
    0.06
    Negative Logits
     Preferences
    -0.78
     Countries
    -0.74
     Extend
    -0.69
     offsets
    -0.69
     Taxes
    -0.69
     Unless
    -0.67
     entails
    -0.67
     IMAGES
    -0.67
     Strategies
    -0.66
     Needs
    -0.66
    POSITIVE LOGITS
     arrested
    1.19
     detained
    1.17
     hospitalized
    1.16
     murdered
    1.10
     killed
    1.10
     stabbed
    1.09
     ejected
    1.08
     abducted
    1.08
     able
    1.08
     assaulted
    1.07
    Activations Density 0.197%

    No Known Activations